Opportunities and Challenges of Applying Artificial Intelligence in Healthcare

Azza Moustafa Fahmy
Theodore Bilharz Research Institute, Giza, Egypt
Correspondence to: hageradel23@yahoo.com

DOI: https://doi.org/10.70389/PJAI.100003

Additional information

• Ethical approval: N/a
• Consent: N/a
• Funding: No industry funding
• Conflicts of interest: N/a
• Author contribution: Azza Moustafa Fahmy – Conceptualization, Writing – original draft, review and editing
• Guarantor: Azza Moustafa Fahmy
• Provenance and peer-review:
  Commissioned and externally peer-reviewed
• Data availability statement: N/a

Keywords: Artificial intelligence, Healthcare, Technologies, Machine learning, Deep learning.

Peer Review
Received: 27 September 2024
Revised: 10 October 2024
Accepted: 10 October 2024
Published: 14 October 2024

Introduction

In order to inform clinical choices based on empirical research and clinical experience, healthcare professionals engage in a dynamic, cyclical process of data collection, analysis, and integration from a variety of sources.¹ Many factors contribute to the stress experienced by healthcare professionals, including long hours, administrative demands, lack of peer support, uncertainty regarding patient care, and emotional strain.² It is especially difficult to adjust to systemic changes because of these considerations.³ These challenges suggest that serious investigation into the various tactics and policies adopted by health systems is necessary, to raise standards on enhancing the satisfaction, productivity, quality, and safety of staff.

Large volumes of data are produced as the healthcare sector rapidly adopts new technologies, but they are not being used enough to prevent disease.⁴ Technological innovations are being used to enhance healthcare research and practice, especially in the areas of robots and artificial intelligence (AI). The goal of the quickly developing computer science subject of AI is to build machines that are capable of activities that normally require human intelligence. AI encompasses a range of methodologies, including natural language processing (NLP), deep learning (DL), and machine learning (ML), and big Language Models are a class of AI algorithms that comprehend, summarize, produce, and forecast new text-based material by utilizing DL techniques and very big datasets.^5,6 The body of research on AI’s potential to improve crucial healthcare procedures and results has grown dramatically over the last 10 years. Personalized therapy, enhanced patient safety, improved clinical processes, and diagnostic support are all possible with AI-based decision support systems. Optimizing patient care and resource management requires a transition from manual to automated systems, with AI, particularly ML, playing a key role.⁷

AI has become a game-changing instrument in healthcare, with potential ways to improve treatment plans, diagnostics, and operational effectiveness. There are many different uses of AI in healthcare, from bettering decision-making and patient outcomes to maximizing system performance.⁸ According to Mahajan et al.⁹, ML can anticipate crucial surgical events and reduce mistakes in diagnosis and therapy. In the fields of personalized medicine, predictive analytics, and medical imaging, advanced ML and DL algorithms have shown impressive results.¹ For instance, AI models are already much better at detecting several types of malignancy and retinal disease using image recognition technology.² It also plays a critical role in enhancing medicine safety by the identification of drug interactions with the use of pre- and postmarketing surveillance.¹⁰ Healthcare professionals might possibly transform patient care by delivering more precise and timely treatments by incorporating AI into clinical decision-making.³

Although much potential exists with healthcare AI integration, the task is really fraught with difficulties. Some key reasons for barring mainstream adoptions for AI-driven decision-making processes are algorithmic bias, data privacy, and lack of transparency.^11,12 AI models are only as good as the data they are trained on since biased training datasets result in biased healthcare outcomes that disproportionally harm disadvantaged populations.¹³ This calls into question the justice and equality of using AI, particularly in critical care settings where reliance on AI advice is essential. Proactive frameworks and regulations that set a high priority on equity, data security, and transparency are needed to address these issues and guarantee the moral and ethical application of AI in healthcare. This review’s objective is to examine the benefits and drawbacks of using AI in the healthcare industry. It specifically looks at how AI technologies—ML in particular—can improve treatment planning, diagnostic accuracy, and the effectiveness of the healthcare system. This paper will also critically examine the main obstacles to AI integration, such as algorithmic bias, data privacy, and legal issues, and offer suggestions for viable approaches to guarantee the efficient, safe, and successful application of AI in healthcare settings.

Opportunities of AI in Healthcare

AI has the potential to improve patient outcomes, increase the accuracy of medical diagnosis, and facilitate the effective use of resources. Healthcare practitioners may use AI-enabled technologies for improved decision-making and outcome improvement through data analytics, pattern recognition, and predictive modeling.¹⁴ Despite the difficulties and moral dilemmas mentioned above, AI offers enormous and far-reaching potential to revolutionize healthcare. The healthcare sector may fully realize AI’s promise to improve patient outcomes, expedite processes, and cut costs by methodically tackling these obstacles with focused policies and integration tactics. The most exciting developments in AI-driven healthcare are outlined in the section that follows. These include AI-enhanced robotic surgery, personalized medication, and predictive analytics. The promise of AI includes clinical decision support systems (CDSSs), robotic surgery, drug development, and operational efficiency—a pretty significant revolution in healthcare delivery.¹⁵ However, there are other obstacles to overcome, including those related to ethics, regulations, and data protection, in order to successfully integrate AI into the present healthcare infrastructure.¹⁶

AI-Driven Diagnostic Tools

When compared to conventional diagnostic techniques, AI offers improved efficiency and accuracy, making it a potent tool in the diagnostic space. Clinicians may now analyze enormous volumes of data with surprising precision by utilizing AI algorithms like ML and DL, which help in the identification of many diseases including cancer, brain problems, and heart disease.¹⁷ A specific class of the DL model that has shown great promise in medical image analysis is the convolutional neural network (CNN), which performs very well in tasks including anomaly detection, segmentation, and picture classification.¹⁸ In addition to decreasing human error, these AI systems expedite diagnosis, enabling earlier identification and more effective treatment planning.¹⁹

Medical imaging is also being transformed by AI diagnostic technologies. AI lessens radiologists’ cognitive burden by automating the examination of radiological images from modalities including MRIs, CT scans, and X-rays, improving diagnosis accuracy and speed.²⁰ According to Li et al.²¹, CNNs have shown great promise in identifying subtle differences in tissue architecture and frequently outperform human specialists in identifying intricate illnesses like breast cancer and interstitial lung disease. This makes it possible for healthcare systems to enhance patient outcomes and diagnostic effectiveness.

Robotic Surgery and Automation

Robotic surgery has changed as a result of the use of AI, which increases accuracy, lowers human error, and improves patient outcomes. Surgeons may now perform complex surgeries with exceptional accuracy, thanks to AI-powered surgical robots that are outfitted with sophisticated algorithms. This results in a considerable reduction in recovery periods and problems.²² These systems help surgeons by optimizing motions and provide real-time monitoring of the surgical field. As a result, they improve safety and lessen surgeon fatigue. Features include motion control, image identification, and haptic feedback.²³ Beyond automating procedures like tissue dissection and suturing, recent advancements in AI-driven robotic surgery now provide improved decision support during surgery. CDSSs with AI capabilities have shown a significant decrease in surgical complications in a variety of procedures, highlighting the function of AI in surgical intervention optimization.²⁴ Further democratization of surgical care means that AI will, for instance, make devices like the AI-GUIDE transform emergency interventions so that nonexperts can safely conduct complex procedures such as central vascular access with unprecedented levels of precision.²⁵

AI-powered simulations are also making major advancements in the field of surgical training. With the application of AI, those simulators can give specific feedback that will enhance intraoperative vision and preoperative planning and result, eventually, in increased patient safety. Although effective in a clinical setting, prohibitively high prices and technological constraints prevent the widespread dissemination of AI-enabled simulators.²⁶ Even though AI has a lot of potential for robotic surgery in the future, there are still obstacles that need to be overcome in order to fully achieve this potential, including high development costs, ethical issues, and regulatory barriers.²⁷

AI in Drug Discovery and Personalized Medicine

The arduous, expensive, and time-consuming procedure that has historically been connected with drug discovery in the biopharmaceutical business has been replaced by AI in the drug development process. AI now makes it possible to rapidly develop new drugs, as it automates key steps in this process: target identification, lead chemical discovery, and clinical trial design.²⁸ AI-driven drug development relies heavily on ML algorithms, DL architectures, and NLP methods to anticipate molecular interactions and therapeutic effectiveness with greater precision.²⁹

AI is especially useful when it comes to drug repurposing, which involves analyzing large databases to find current medications that could be useful for novel therapeutic uses. This strategy has shown to be essential during the COVID-19 pandemic, as AI-driven techniques have sped up the identification of possible SARS-CoV-2 and its variant therapies.³⁰ Pharmaceutical corporations have benefited financially from the use of AI in medication repurposing and development, which has also sped up the time it takes to produce new drugs and given patients faster access to potentially life-saving therapies. AI is also revolutionizing personalized medicine by enabling researchers and medical professionals to customize patient care, according to their individual genetic composition, biomarkers, and medical background. AI algorithms can forecast the best courses of action for specific patients by utilizing multiomics data and electronic health records (EHRs). This opens the door to tailored therapeutics for difficult-to-treat conditions including Alzheimer’s and cancer.²⁹ AI is predicted to play an increasingly larger role in drug development and personalized medicine as technology develops, presenting previously unheard-of chances for therapeutic innovation and better patient outcomes.

Enhancing Operational Efficiency

The effectiveness of hospitals and patient care have significantly improved as a result of the introduction of AI into healthcare operations. AI-driven solutions enhance workflow and save expenses by streamlining administrative duties including scheduling, resource allocation, and inventory management.³¹ In order to effectively manage hospital capacities, such as bed availability and staff allocation, and better meet patient demand, predictive models and ML algorithms analyze large datasets to enhance decisions.³² Improved decision-making skills, a lower chance of human mistakes, and automated systems that maximize operational efficiency are other advantages for hospitals that use AI. To guarantee responsible AI deployment, however, ethical considerations around data privacy and regulatory compliance continue to be crucial and must be carefully considered in tandem with these breakthroughs.³³

Challenges of AI in Healthcare

AI in healthcare has the potential to completely transform a number of areas of medical practice, including administrative work, operational efficiency, and personalized treatment plans.³⁴ Notwithstanding these encouraging developments, the application of AI in healthcare has a number of obstacles that need to be carefully considered in order to guarantee its safe, moral, and efficient use. The main barriers to the smooth integration of AI in healthcare systems are examined in this section. These include worries about data security and privacy, ethical and legal challenges, bias in AI algorithms, integration with current infrastructure, and patient and provider acceptability.

Data Privacy and Security Concerns

Since health data, particularly patient information, are very sensitive and subject to stringent regulations, one cannot help but raise severe concerns about privacy and security. The ethical and legal frameworks that should govern the use of data have gaps as a result of the quick expansion of AI in the healthcare industry. The growing size of datasets poses a risk to privacy since they contain both unprotected and protected health information as well as data from smart devices.³⁵ Good governance practices in the protection of AI models against cyberattacks and data breaches will go a long way in safeguarding patient data. Although Federated Learning enables training AI models in a decentralized data fashion, reducing the risk of disclosure of sensitive information and ensuring compliance with all requirements around privacy, blockchain, among other emerging technologies, ensures tamper-proof data records.³⁶

Regulatory and Ethical Issues

It is crucial to remember that AI health legislation is still in its infancy and that the current frameworks are unable to handle the unique challenges that AI presents, not the least of which are those pertaining to patient permission, responsibility, and accountability. Given the “black box” nature of many AI systems, assigning culpability gets challenging as AI technologies such as sophisticated diagnostic models and autonomous surgical robots advance.³⁷ Because of the serious concerns raised by stakeholders, explainable AI (XAI) has to be developed in order to guarantee ethical norms, transparency, and confidence in healthcare AI applications.³⁸ Different stakeholders—from developers to healthcare professionals—have different needs when it comes to explanations; therefore, the demand for XAI is not universal. Building confidence in AI systems requires attending to these demands. There are supervision holes in AI since current laws, such as those from the Food and Drug Administration (FDA), mostly focus on traditional medical devices. Standardized evaluation procedures are made more difficult by ML algorithms that adjust to fresh data.³⁹ Clear legal frameworks that define the roles and duties of healthcare professionals, AI developers, and institutions are needed to solve these concerns. Additionally, upholding patient trust and encouraging ethical AI use in healthcare will depend on making sure AI systems are understandable to a variety of stakeholders.

Bias in AI Algorithms

In healthcare, biased AI algorithms may result in unequal treatment among patients from diverse demographics due to unrepresentative training datasets. Due to a lack of diverse training information, AI models in dermatology have clearly manifested shortcomings in their diagnostic capabilities in patients with darker skin tones. The research underlined that these models are less representative of patients with darker skin tones, emphasizing the critical need for representative datasets.⁴⁰ Healthcare disparities are exacerbated in other domains, such as medical imaging, where AI models trained on homogeneous datasets do not generalize well to varied populations.⁴¹ The only possible way to handle bias in AI is by increasing the variety of datasets used for training models. This way, researchers can be assured that their AI models work fairly across different patient groups if they have more detailed demographic information, such as race and geography.⁴²

Integration with Existing Infrastructure

It is difficult to integrate AI into healthcare systems because of the constraints of the current infrastructure. One major barrier to better integration of AI is the inability of systems to talk to each other and apply specified formats of data. Even though healthcare systems generate huge volumes of data, it remains quite cumbersome and difficult to transfer such data across different providers efficiently and securely. Using AI to improve medical outcomes requires semantic interoperability, which allows systems to properly communicate and understand data .⁴³ Healthcare systems need to emphasize safe data-sharing protocols to enable interoperability, engage in staff retraining, and concentrate on phased adoption to guarantee the effective integration of AI interoperability.⁴⁴

Patient and Provider Acceptance

Obstacles can be related to both patients and the healthcare provider’s end. This makes health workers concerned with the use of AI in the industry, as they feel it is going to cut their autonomy and make them rely on AI technologies in clinical processes. Clinicians’ suspicion and skepticism of AI are growing due to worries about its capacity to make trustworthy and correct choices.^45,46 Patients have similar apprehensions, such as treatment depersonalization and AI reliability. They also have doubt because they do not know whether AI is sensitive enough to be helpful for people with special needs like them. Furthermore, the absence of uniform criteria for the ethical use of AI worries patients as well as clinicians, especially in regard to data security and privacy.^47,48 For AI to successfully integrate into healthcare, building trust is essential. This may be done by addressing these concerns through ethical frameworks, education, and openness.

Key Technologies and Techniques

AI in healthcare depends on different key technologies and methodologies for its development in order to enhance the industry. Through the interpretation of complex medical data, ML and DL have been able to make predictive medicine, personalization of therapy, and accuracy in diagnosis possible. NLP increases narrative information processing in medical records through the transformation of unstructured data to structured representation for better decision-making done by Iroju and Olaleke.⁴⁹ Computer vision (CV) has been instrumental in enabling medical imagery to be transformed to allow much more detailed picture interpretation in certain specialties such as pathology and cardiology.⁵⁰ Numerous AI technologies have also greatly enhanced telemedicine and remote care, offering previously unheard-of possibilities for patient monitoring and healthcare access in resource-constrained settings.

ML and DL

While ML and DL have caused a shift in the health domain, they now enable data-driven insights that lead to improved decision-making. Unlike traditional programming, ML deploys statistical models to provide outputs from large datasets without explicit human supervision; it needs feature extraction approaches in order to extract the data patterns.⁵⁰ As a subset of ML, DL automates this process by using neural networks to gradually convert raw data into abstract representations. This makes DL useful for intricate applications like medical image analysis and illness detection.⁵⁰ Furthermore, to enhance predictive modeling across illnesses, instruments such as the “deep patient” leverage data from EHRs.⁵¹ According to Kassahun et al.⁵², these are technologies that provide more precise diagnoses and customized therapies through the extraction of insights from large databases.

NLP

Healthcare is evolving, thanks to NLP, which enables AI systems to examine enormous volumes of unstructured medical text produced by PGHD and EHRs. While free text in EHRs was hard to parse with the traditional systems, this has been made quite easy through new neural network-based NLP techniques for extracting key data for categorization, prediction, and decision-making activities.⁵³ According to pipelines utilizing Named Entity Recognition and medical ontologies like SNOMED CT, NLP also improves symptom analysis and medication adherence tracking from PGHD.⁵⁴ Results research is aided by tools such as Canary, which let nontechnical users to extract actionable insights from provider notes.⁵⁵ Advanced NLP models, which include BERT and GPT, have contributed to the even better processing of complex clinical narratives while in turn reducing administrative burdens among health professionals, improving the accuracy of documentation, and enhancing symptom extraction.⁵⁶

CV in Medical Imaging

CV applications in medical imaging have been revolutionized by recent advances in DL, especially CNNs. These advancements have improved tasks like object recognition, segmentation across modalities like MRIs, CT scans, and X-rays, and classification.⁵⁷ Crowdsourcing provides a way to handle large-scale data annotation in spite of obstacles like few datasets and expensive annotation fees.⁵⁸ High diagnostic accuracy has been demonstrated by AI-powered CV systems; examples include the use of CNNs to identify pneumonia in chest X-rays⁵⁹ and U-Net to automate the segmentation of organs and tumors.⁶⁰ New models such as RNNs and GANs are boosting time-series analysis and synthetic picture synthesis, which helps with illness progression comprehension and diagnosis.⁶¹ Through real-time analysis and AI-guided treatments, the integration of CV technology into minimally invasive surgical techniques has improved accuracy.⁶²

AI in Telemedicine and Remote Care

With the COVID-19 pandemic in particular, AI has accelerated its revolution in telemedicine and remote treatment. The use of virtual consultations, remote diagnostics, and EHR integration has improved clinical procedures and monitored patient diagnosis accuracy.⁶³ Wearable sensors and AI-powered telemedicine systems monitor chronic diseases in real time, thus providing predictive analytics and data to facilitate timely therapies among at-risk patients.⁶⁴ Furthermore, continuous Remote Patient Monitoring is made possible by AI and Internet of Medical Things technologies, which also minimize hospital readmissions and improve therapy.⁶⁵ However, there are a couple of issues that are yet to be solved regarding the integration: those being ethics, regulations, and training. In the near future, AI in telemedicine has good prospects that will further advance remote medical delivery by strengthening its resilience and efficiency.⁶⁶

Discussion

Balancing Innovation with Ethical Standards

AI is beneficial to the healthcare industry in many ways, but its use has to walk a fine line between originality and stern morals. It becomes very morally essential in AI since it is incredibly capable of processing massive quantities of sensitive medical data and making conclusions that directly affect patient health. One important concern here is data privacy, as AI systems depend significantly on giant datasets-most times situated in cloud-based systems, which are still quite vulnerable to breaches and misuse.⁶ Furthermore, because AI models are only as good as the data they are trained on, algorithmic bias is an ongoing problem. If these datasets are not diverse and representative, the models might perpetuate healthcare inequities by underestimating the requirements of specific populations, particularly minorities.¹³ There is growing consensus that in order to ensure that every demographic group is treated fairly and equally, AI systems must be developed and evaluated, utilizing a variety of datasets.¹²

Another ethical issue related to AI decision-making is transparency. Given the complexity of their decision-making processes, AI models—very specialized ones based on DL—are sometimes described as “black boxes,” as they are incomprehensible to health personnel.¹¹ This lack of transparency might undermine trust in decisions made by AI, especially when lives are at stake. As a result, the creation of XAI, which attempts to increase trust and responsibility by making AI decisions simpler for customers to understand, is receiving more attention.⁶⁷ The subsequent case studies offer useful insights into the application of AI in healthcare by demonstrating how top organizations and authorities are handling these moral dilemmas:

Case Study: Reducing Algorithmic Bias in Radiology AI

By reorganizing training datasets to incorporate a greater variety of demographic data, Massachusetts General Hospital addressed algorithmic bias in its radiology AI models, improving diagnostic accuracy across various skin tones.⁴⁰ Initially, because there was less diversity in the training data, the AI models performed less accurately for patients with darker skin tones. This project emphasizes how crucial it is to use representative and diverse datasets in clinical AI applications to guarantee just and equitable results.

Case Study: FDA’s Regulatory Framework for AI-Based Diagnostic Tools

The Software as a Medical Device (SaMD) framework was developed by the U.S. FDA in 2020 to facilitate the approval of AI-based diagnostic tools in a more flexible manner. With the help of this framework, AI systems may learn and update continually without needing to be reapproved, allowing for quicker adaptation to new data. The FDA-approved IDx-DR AI system for diagnosing diabetic retinopathy is one such example; since its clinical deployment in 2018, it has shown accuracy on par with human physicians.¹³

Global Ethical Standards: The European Union’s Artificial Intelligence Act

Internationally, the Artificial Intelligence Act proposed by the European Union establishes a precedent by grading AI applications based on risk, requiring more stringent assessments for high-risk healthcare systems. By addressing ethical issues like algorithmic bias and transparency, this framework makes sure AI systems are subject to ongoing oversight and adhere to strict safety and accountability guidelines.³⁷ Such all-encompassing rules are essential for the moral use of AI in healthcare, promoting accountability and trust in a variety of healthcare settings. These moral questions highlight the necessity for extensive legal frameworks that guarantee the ethical, responsible, and transparent development and application of AI technology. By establishing such frameworks, patients and healthcare professionals can feel more confident and trusting of one another while addressing concerns like algorithmic bias and data privacy.

Recommendations for AI Integration in Healthcare

AI technology integration in the healthcare industry necessitates a methodical strategy that takes into account both pragmatic implementation techniques and regulatory considerations. The purpose of the following guidelines is to help healthcare professionals and legislators make sure that the use of AI improves patient care, fosters transparency, and upholds ethical norms.

Recommendations for Policymakers

Establish Comprehensive Regulatory Frameworks:

Create uniform policies and risk-based categorizations for AI systems, such as the Artificial Intelligence Act of the European Union, which assigns a risk rating to AI applications and requires stringent assessments for high-risk healthcare systems.³⁷ This approach makes sure that before AI applications are used in clinical settings, they adhere to safety, efficacy, and ethical requirements.

Promote Transparency and Accountability:

Establish mandates for required explainability and transparency evaluations, requiring AI developers to reveal the capabilities, constraints, and methods of decision-making underlying their models. The FDA’s SaMD framework, which encourages ongoing education and openness in AI tools used for diagnosis, has put out such things.

Support Data Sharing and Interoperability:

Adopt procedures that protect patient privacy and enable safe data sharing between healthcare organizations. The quality and diversity of datasets used to train AI models can be increased by promoting cooperation through standardized data-sharing protocols, which reduces biases and enhances model performance.⁴⁰

Continuous Monitoring and Compliance:

To guarantee that AI systems continue to function safely and morally as they advance, put in place procedures for continuous observation and postmarket surveillance. The FDA’s flexible clearance procedure for AI tools demonstrates how rules can change as technology advances by allowing modifications without requiring new approval.¹³

Recommendations for Healthcare Practitioners

Adopt a Phased Implementation Strategy:

When introducing AI technologies, start with low-risk uses to gauge how they affect clinical workflows. Using this method enables healthcare organizations to evaluate possible dangers, provides personnel with appropriate training, and makes the required modifications prior to implementing more sophisticated AI applications.⁶⁸

Invest in Comprehensive Staff Training and Education:

To acquaint healthcare workers with AI technologies, offer focused training programs that emphasize comprehending AI outputs, identifying potential biases, and incorporating AI recommendations into clinical decision-making.⁴⁰ In order to reduce resistance and increase confidence when utilizing AI, proper training is crucial.

Create Interdisciplinary AI Review Committees:

Form multidisciplinary committees comprising data scientists, doctors, and ethicists to supervise the implementation and functionality of AI systems. These committees are capable of analyzing AI performance, evaluating ethical implications, and resolving any issues that may come up during implementation.³⁷

Implement Rigorous Data Validation Protocols:

To guarantee that AI models are trained on reliable, impartial data, implement strong data validation processes. In order to identify and address biases and guarantee that AI tools provide fair and dependable suggestions, it is imperative to conduct routine evaluations of the model’s performance and output consistency.⁴⁰

Encourage Continuous Feedback and Iterative Improvement:

To find areas where AI applications need to be improved, encourage a culture of ongoing input from patients and healthcare professionals. AI tools should be updated frequently in response to feedback from the actual world to keep them current, useful, and in line with therapeutic objectives.⁶⁸

The Future of AI in Healthcare: Emerging Trends and Technologies

With a number of new developments and technologies ready to completely transform the healthcare industry, AI has a bright future. For instance, by anticipating disease outbreaks and identifying high-risk individuals for preventative treatments, AI-driven predictive analytics is anticipated to play a significant role in the population health management.⁶⁸ Furthermore, by examining international datasets, AI-driven platforms like BlueDot have shown the ability to anticipate and control infectious disease epidemics. BlueDot demonstrated the importance of AI in improving population health management by identifying and notifying health authorities about the COVID-19 outbreak in Wuhan days before it was formally reported.⁶⁸Another emerging trend is personalized medicine, which makes use of AI to examine patients’ genetic profiles and create customized treatment regimens, especially in areas like cancer.⁶⁹ By evaluating patients’ genetic profiles and clinical records to suggest customized treatment options, AI systems like IBM’s Watson for Oncology are transforming personalized medicine, especially in the field of oncology.²⁹

Robotic surgery, where AI-powered robots help surgeons carry out intricate procedures with more precision, is another rapidly growing field. Robotics has the potential to improve surgical outcomes and shorten recovery periods; systems such as the da Vinci Surgical System have already shown this.⁷⁰ AI is also making advances in the field of diagnosis. According to Gulshan et al. ⁷¹, image recognition algorithms have demonstrated the ability to identify conditions including cancer, diabetic retinopathy, and cardiovascular disorders at a level that is on par with or even higher than that of human professionals. Chatbots and virtual health assistants with AI capabilities are becoming essential tools for improving patient access and engagement with healthcare. According to Alowais et al.³⁴, these technologies can help with appointment scheduling, respond to patient questions in real time, and provide health recommendations based on symptoms indicated by the patient. Furthermore, AI is being utilized more and more in CDSSs, where it evaluates medical literature, analyzes patient data, and recommends evidence-based treatment alternatives to assist doctors in making better judgments.⁷² AI’s significance in healthcare will only grow as these technologies advance, providing creative answers to enduring problems and enhancing patient outcomes overall.

Policy Recommendations and Regulatory Considerations

In-depth legislative frameworks and regulations that handle the particular difficulties presented by AI technologies are necessary in order to fully realize the promise of AI in healthcare. The creation of precise policies for data security and privacy is a top priority. To prevent breaches and illegal use of patient information, governments and regulatory agencies need to impose strict data protection procedures.⁷³ This is especially crucial because AI systems depend on massive, cloud-based databases. The evaluation and certification of AI algorithms used in healthcare is another important regulatory problem. The responsibility of ensuring that AI systems deployed in clinical settings fulfill high criteria of accuracy, safety, and efficacy is falling increasingly on regulatory agencies like the European Medicines Agency and the U.S. FDA.⁷⁴ Nonetheless, a lot of AI technologies are still in their infancy, and laws and regulations frequently keep pace with rapid technical development. Regulations must be revised to reflect the quick speed at which AI is developing while maintaining patient safety and moral application.

Although there is a lot of promise for integrating AI technologies into healthcare, our legal frameworks need to be reevaluated. The big question, however, is how the traditional concepts of liability must be adapted to the peculiar and unpredictable nature of AI, or how the focus needs to shift from liability to reconciliation. Subsequent discourses and inquiries should go through these elaborate landscapes and strive for solutions that ensure progress and conservation.⁷⁵ By encouraging openness in AI algorithms, policies should also support the deployment of XAI systems, which allow patients and healthcare providers to understand the logic behind AI-generated decisions.⁷⁶

Conclusion

With AI being introduced into the healthcare sector, several benefits could be identified, including the enhancement of treatment and diagnosis and improvement in operational efficiency. AI technologies include robotic surgery, DL, and ML; these, in turn, are enhancing predictive analytics, CDSSs, and customized medicine. Yet, there is a set of open issues, but most importantly, the ones related to algorithmic bias, data privacy, and transparency in decision-making processes by AI systems. Its regulation in regard to healthcare is yet to be decided upon, which means the laws have to be designed in such a way that they could resolve these issues while being able to promote innovation.

References

1. Huo M, Zhao B, Li Y, Li J. Evidence-based practice dynamic capabilities: a concept derivation and analysis. Ann Transl Med [Internet]. 2022;10(1):22. Available from: https://atm.amegroups.com/article/view/87242/html
https://doi.org/10.21037/atm-21-6506
 
2. Rink LC, Oyesanya TO, Adair KC, Humphreys JC, Silva SG, Sexton JB. Stressors among healthcare workers: a summative content analysis. Glob Qual Nurs Res [Internet]. 2023;10:233339362311611. Available from: http://journals.sagepub.com/doi/10.1177/23333936231161127
https://doi.org/10.1177/23333936231161127
 
3. Murthy VH. Confronting health worker burnout and well-being. N Engl J Med [Internet]. 2022;387(7):577-9. Available from: http://www.nejm.org/doi/10.1056/NEJMp2207252
https://doi.org/10.1056/NEJMp2207252
 
4. Olaronke I, Oluwaseun O. Big data in healthcare: prospects, challenges and resolutions. In: 2016 Future Technologies Conference (FTC) [Internet]. IEEE; 2016. pp. 1152-7. Available from: http://ieeexplore.ieee.org/document/7821747/
https://doi.org/10.1109/FTC.2016.7821747
 
5. Suleimenov IE, Vitulyova YS, Bakirov AS, Gabrielyan OA. Artificial intelligence. In: Proceedings of the 2020 6th International Conference on Computer and Technology Applications [Internet]. New York, NY, USA: ACM; 2020. pp. 22-5. Available from: https://dl.acm.org/doi/10.1145/3397125.3397141
https://doi.org/10.1145/3397125.3397141
 
6. Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Futur Healthc J [Internet]. 2019;6(2):94-8. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2514664524010592
https://doi.org/10.7861/futurehosp.6-2-94
 
7. Alanazi A. Using machine learning for healthcare challenges and opportunities. Informatics Med Unlocked [Internet]. 2022;30:100924. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2352914822000739
https://doi.org/10.1016/j.imu.2022.100924
 
8. Chen M, Decary M. Artificial intelligence in healthcare: an essential guide for health leaders. Healthc Manag Forum [Internet]. 2020;33(1):10-8. Available from: http://journals.sagepub.com/doi/10.1177/0840470419873123
https://doi.org/10.1177/0840470419873123
 
9. Mahajan A, Esper SA, Cole DJ, Fleisher LA. Anesthesiologists’ role in value-based perioperative care and healthcare transformation. Anesthesiology [Internet]. 2021;134(4):526-40. Available from: https://pubs.asahq.org/anesthesiology/article/134/4/526/115421/Anesthesiologists- Role-in-Value-based
https://doi.org/10.1097/ALN.0000000000003717
 
10. Liu N, Chen CB, Kumara S. Semi-supervised learning algorithm for identifying high-priority drug-drug interactions through adverse event reports. IEEE J Biomed Heal Informatics [Internet]. 2020;24(1):57-68. Available from: https://ieeexplore.ieee.org/document/8786248/
https://doi.org/10.1109/JBHI.2019.2932740
 
11. Waring J, Lindvall C, Umeton R. Automated machine learning: review of the state-of-the- art and opportunities for healthcare. Artif Intell Med. 2020;104:101822. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0933365719310437
https://doi.org/10.1016/j.artmed.2020.101822
 
12. Char DS, Shah NH, Magnus D. Implementing machine learning in health care-addressing ethical challenges. N Engl J Med [Internet]. 2018;378(11):981-3. Available from: http://www.nejm.org/doi/10.1056/NEJMp1714229
https://doi.org/10.1056/NEJMp1714229
 
13. Obermeyer Z, Emanuel EJ. Predicting the future-big data, machine learning, and clinical medicine. N Engl J Med [Internet]. 2016;375(13):1216-9. Available from: http://www.nejm.org/doi/10.1056/NEJMp1606181
https://doi.org/10.1056/NEJMp1606181
 
14. Patil S, Shankar H. Transforming healthcare: harnessing the power of AI in the modern era. Int J Multidiscip Sci Arts [Internet]. 2023;2(2):60-70. Available from: https://jurnal.itscience.org/index.php/ijmdsa/article/view/2513
https://doi.org/10.47709/ijmdsa.v2i1.2513
 
15. Reddy S, Fox J, Purohit MP. Artificial intelligence-enabled healthcare delivery. J R Soc Med [Internet]. 2019;112(1):22-8. Available from: http://journals.sagepub.com/doi/10.1177/0141076818815510
https://doi.org/10.1177/0141076818815510
 
16. Dicuonzo G, Donofrio F, Fusco A, Shini M. Healthcare system: moving forward with artificial intelligence. Technovation [Internet]. 2023;120:102510. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0166497222000578
https://doi.org/10.1016/j.technovation.2022.102510
 
17. Kaur S, Singla J, Nkenyereye L, Jha S, Prashar D, Joshi GP, et al. Medical diagnostic systems using artificial intelligence (AI) algorithms: principles and perspectives. IEEE Access [Internet]. 2020;8:228049-69. Available from: https://ieeexplore.ieee.org/document/9279211/
https://doi.org/10.1109/ACCESS.2020.3042273
 
18. Sarvamangala DR, Kulkarni RV. Convolutional neural networks in medical image understanding: a survey. Evol Intell [Internet]. 2022;15(1):1-22. Available from: https://link.springer.com/10.1007/s12065-020-00540-3
https://doi.org/10.1007/s12065-020-00540-3
 
19. Ghaffar Nia N, Kaplanoglu E, Nasab A. Evaluation of artificial intelligence techniques in disease diagnosis and prediction. Discov Artif Intell [Internet]. 2023;3(1):5. Available from: https://link.springer.com/10.1007/s44163-023-00049-5
https://doi.org/10.1007/s44163-023-00049-5
 
20. Tajbakhsh N, Shin JY, Gurudu SR, Hurst RT, Kendall CB, Gotway MB, et al. Convolutional neural networks for medical image analysis: full training or fine tuning? IEEE Trans Med Imaging [Internet]. 2016;35(5):1299-312. Available from: http://ieeexplore.ieee.org/document/7426826/
https://doi.org/10.1109/TMI.2016.2535302
 
21. Li Q, Cai W, Wang X, Zhou Y, Feng DD, Chen M. Medical image classification with convolutional neural network. In: 2014 13th International Conference on Control Automation Robotics & Vision (ICARCV) [Internet]. IEEE; 2014. p. 844-8. Available from: http://ieeexplore.ieee.org/document/7064414/
https://doi.org/10.1109/ICARCV.2014.7064414
 
22. Abbasi N, Hussain HK. Integration of artificial intelligence and smart technology: AI-driven robotics in surgery: precision and efficiency. J Artif Intell Gen Sci ISSN3006-4023 [Internet]. 2024;5(1):381-90. Available from: https://ojs.boulibrary.com/index.php/JAIGS/article/view/207
https://doi.org/10.60087/jaigs.v5i1.207
 
23. Iftikhar M, Saqib M, Zareen M, Mumtaz H. Artificial intelligence: revolutionizing robotic surgery: review. Ann Med Surg [Internet]. 2024;86(9):5401-9. Available from: https://journals.lww.com/10.1097/MS9.0000000000002426
https://doi.org/10.1097/MS9.0000000000002426
 
24. Shahi A, Bajaj G, GolharSathawane R, Mendhe D, Dogra A. Integrating robot-assisted surgery and AI for improved healthcare outcomes. In: 2024 Ninth International Conference on Science Technology Engineering and Mathematics (ICONSTEM) [Internet]. IEEE; 2024. p. 1-5. Available from: https://ieeexplore.ieee.org/document/10568646/
https://doi.org/10.1109/ICONSTEM60960.2024.10568646
 
25. Brattain L, Pierce T, Gjesteby L, Johnson M, DeLosa N, Werblin J, et al. AI-enabled, ultrasound-guided handheld robotic device for femoral vascular access. Biosensors [Internet]. 2021;11(12):522. Available from: https://www.mdpi.com/2079- 6374/11/12/522
https://doi.org/10.3390/bios11120522
 
26. Park JJ, Tiefenbach J, Demetriades AK. The role of artificial intelligence in surgical simulation. Front Med Technol [Internet]. 2022;4. Available from: https://www.frontiersin.org/articles/10.3389/fmedt.2022.1076755/full
https://doi.org/10.3389/fmedt.2022.1076755
 
27. Loftus TJ, Altieri MS, Balch JA, Abbott KL, Choi J, Marwaha JS, et al. Artificial intelligence- enabled decision support in surgery. Ann Surg [Internet]. 2023;278(1):51-8. Available from: https://journals.lww.com/10.1097/SLA.0000000000005853
https://doi.org/10.1097/SLA.0000000000005853
 
28. Tripathi A, Misra K, Dhanuka R, Singh JP. Artificial intelligence in accelerating drug discovery and development. Recent Pat Biotechnol [Internet]. 2023;17(1):9-23. Available from: https://www.eurekaselect.com/207223/article
https://doi.org/10.2174/1872208316666220802151129
 
29. Huang D, Yang M, Wen X, Xia S, Yuan B. AI-driven drug discovery: accelerating the development of novel therapeutics in biopharmaceuticals. J Knowl Learn Sci Technol ISSN 2959-6386 [Internet]. 2024;3(3):206-24. Available from: https://jklst.org/index.php/home/article/view/240
https://doi.org/10.60087/jklst.vol3.n3.p.206-224
 
30. Farghali H, Kutinová Canová N, Arora M. The potential applications of artificial intelligence in drug discovery and development. Physiol Res [Internet]. 2021;S715-22. Available from: https://www.biomed.cas.cz/physiolres/pdf/2021/70_S715.pdf
https://doi.org/10.33549/physiolres.934765
 
31. Kumar D. AI- driven automation in administrative processes: enhancing efficiency and accuracy. Int J Eng Sci Humanit [Internet]. 2024;14(Special Issue 1):256-65. Available from: https://ijeshonline.com/index.php/ijesh/article/view/49
https://doi.org/10.62904/qg004437
 
32. Ordu M, Demir E, Tofallis C, Gunal MM. A novel healthcare resource allocation decision support tool: a forecasting-simulation-optimization approach. J Oper Res Soc [Internet]. 2021;72(3):485-500. Available from: https://www.tandfonline.com/doi/full/10.1080/01605682.2019.1700186
https://doi.org/10.1080/01605682.2019.1700186
 
33. Maleki Varnosfaderani S, Forouzanfar M. The role of AI in hospitals and clinics: transforming healthcare in the 21st century. Bioengineering [Internet]. 2024;11(4):337. Available from: https://www.mdpi.com/2306-5354/11/4/337
https://doi.org/10.3390/bioengineering11040337
 
34. Alowais SA, Alghamdi SS, Alsuhebany N, Alqahtani T, Alshaya AI, Almohareb SN, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ [Internet]. 2023;23(1):689. Available from: https://bmcmededuc.biomedcentral.com/articles/10.1186/s12909-023-04698-z
https://doi.org/10.1186/s12909-023-04698-z
 
35. Yadav N, Pandey S, Gupta A, Dudani P, Gupta S, Rangarajan K. Data privacy in healthcare: in the era of artificial intelligence. Indian Dermatol Online J [Internet]. 2023;14(6):788-92. Available from: https://journals.lww.com/10.4103/idoj.idoj_543_23
https://doi.org/10.4103/idoj.idoj_543_23
 
36. Coppieters Y, Levêque A. Ethics, privacy and the legal framework governing medical data: opportunities or threats for biomedical and public health research? Arch Public Heal [Internet]. 2013;71(1):15. Available from: https://archpublichealth.biomedcentral.com/articles/10.1186/0778-7367-71-15
https://doi.org/10.1186/0778-7367-71-15
 
37. O’Sullivan S, Nevejans N, Allen C, Blyth A, Leonard S, Pagallo U, et al. Legal, regulatory, and ethical frameworks for development of standards in artificial intelligence (AI) and autonomous robotic surgery. Int J Med Robot Comput Assist Surg [Internet]. 2019;15(1). Available from: https://onlinelibrary.wiley.com/doi/10.1002/rcs.1968
https://doi.org/10.1002/rcs.1968
 
38. Gerlings J, Jensen MS, Shollo A. Explainable AI. But explainable to whom? An exploratory case study of xAI in healthcare. 2022. p. 169-98. Available from: https://link.springer.com/10.1007/978-3-030-83620-7_7
https://doi.org/10.1007/978-3-030-83620-7_7
 
39. Mennella C, Maniscalco U, De Pietro G, Esposito M. Ethical and regulatory challenges of AI technologies in healthcare: a narrative review. Heliyon [Internet]. 2024;10(4):e26297. Available from: https://linkinghub.elsevier.com/retrieve/pii/S2405844024023284
https://doi.org/10.1016/j.heliyon.2024.e26297
 
40. Daneshjou R, Vodrahalli K, Novoa RA, Jenkins M, Liang W, Rotemberg V, et al. Disparities in dermatology AI performance on a diverse, curated clinical image set. Sci Adv [Internet]. 2022;8(32). Available from: https://www.science.org/doi/10.1126/sciadv.abq6147
https://doi.org/10.1126/sciadv.abq6147
 
41. Tripathi S, Gabriel K, Dheer S, Parajuli A, Augustin AI, Elahi A, et al. Understanding biases and disparities in radiology AI datasets: a review. J Am Coll Radiol [Internet]. 2023;20(9):836-41. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1546144023005070
https://doi.org/10.1016/j.jacr.2023.06.015
 
42. Kamulegeya L, Bwanika J, Okello M, Rusoke D, Nassiwa F, Lubega W, et al. Using artificial intelligence on dermatology conditions in Uganda: a case for diversity in training data sets for machine learning. Afr Health Sci [Internet]. 2023;23(2):753-63. Available from: https://www.ajol.info/index.php/ahs/article/view/250826
https://doi.org/10.4314/ahs.v23i2.86
 
43. Sreenivasan M, Chacko AM. Interoperability issues in EHR systems: research directions. In: Data Analytics in Biomedical Engineering and Healthcare [Internet]. 2021;(pp. 13-28). Elsevier. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780128193143000021
https://doi.org/10.1016/B978-0-12-819314-3.00002-1
 
44. Williams E, Kienast M, Medawar E, Reinelt J, Merola A, Klopfenstein SAI, et al. A standardized clinical data harmonization pipeline for scalable AI application deployment (FHIR-DHP): validation and usability study. JMIR Med Informatics [Internet]. 2023;11:e43847. Available from: https://medinform.jmir.org/2023/1/e43847
https://doi.org/10.2196/43847
 
45. Arnold MH. Teasing out artificial intelligence in medicine: an ethical critique of artificial intelligence and machine learning in medicine. J Bioeth Inq [Internet]. 2021;18(1):121-39. Available from: https://link.springer.com/10.1007/s11673-020-10080-1
https://doi.org/10.1007/s11673-020-10080-1
 
46. Liyanage H, Liaw ST, Jonnagaddala J, Schreiber R, Kuziemsky C, Terry AL, et al. Artificial intelligence in primary health care: perceptions, issues, and challenges. Yearb Med Inform [Internet]. 2019;28(01):041-6. Available from: http://www.thieme- connect.de/DOI/DOI?10.1055/s-0039-1677901
https://doi.org/10.1055/s-0039-1677901
 
47. Husnain A, Rasool S, Saeed A, Yousaf Gill A, Khawar Hussain H. AI’S healing touch: examining machine learning’s transformative effects on healthcare. J World Sci [Internet]. 2023;2(10):1681-95. Available from: https://jws.rivierapublishing.id/index.php/jws/article/view/448
https://doi.org/10.58344/jws.v2i10.448
 
48. khan B, Fatima H, Qureshi A, Kumar S, Hanan A, Hussain J, et al. Drawbacks of artificial intelligence and their potential solutions in the healthcare sector. Biomed Mater Devices [Internet]. 2023;1(2):731-8. Available from: https://link.springer.com/10.1007/s44174-023-00063-2
https://doi.org/10.1007/s44174-023-00063-2
 
49. Iroju OG, Olaleke JO. A systematic review of natural language processing in healthcare. Int J Inf Technol Comput Sci [Internet]. 2015;7(8):44-50. Available from: http://www.mecs-press.org/ijitcs/ijitcs-v7-n8/v7n8-7.html
https://doi.org/10.5815/ijitcs.2015.08.07
 
50. Esteva A, Chou K, Yeung S, Naik N, Madani A, Mottaghi A, et al. Deep learning-enabled medical computer vision. npj Digit Med [Internet]. 2021;4(1):5. Available from: https://www.nature.com/articles/s41746-020-00376-2
https://doi.org/10.1038/s41746-020-00376-2
 
51. Miotto R, Li L, Kidd BA, Dudley JT. Deep patient: an unsupervised representation to predict the future of patients from the electronic health records. Sci Rep [Internet]. 2016;6(1):26094. Available from: https://www.nature.com/articles/srep26094
https://doi.org/10.1038/srep26094
 
52. Kassahun Y, Yu B, Tibebu AT, Stoyanov D, Giannarou S, Metzen JH, et al. Surgical robotics beyond enhanced dexterity instrumentation: a survey of machine learning techniques and their role in intelligent and autonomous surgical actions. Int J Comput Assist Radiol Surg [Internet]. 2016;11(4):553-68. Available from: http://link.springer.com/10.1007/s11548-015-1305-z
https://doi.org/10.1007/s11548-015-1305-z
 
53. Li I, Pan J, Goldwasser J, Verma N, Wong WP, Nuzumlalı MY, et al. Neural natural language processing for unstructured data in electronic health records: a review. Comput Sci Rev [Internet]. 2022;46:100511. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1574013722000454
https://doi.org/10.1016/j.cosrev.2022.100511
 
54. Sezgin E, Hussain SA, Rust S, Huang Y. Extracting medical information from free-text and unstructured patient-generated health data using natural language processing methods: feasibility study with real-world data. JMIR Form Res [Internet]. 2023;7:e43014. Available from: https://formative.jmir.org/2023/1/e43014
https://doi.org/10.2196/43014
 
55. Malmasi S, Hosomura N, Chang LS, Brown CJ, Skentzos S, Turchin A. Extracting healthcare quality information from unstructured Data. AMIA. Annu Symp Proceedings AMIA Symp [Internet]. 2017;2017:1243-52. Available from: http://www.ncbi.nlm.nih.gov/pubmed/29854193
 
56. Hussain SA, Sezgin E, Krivchenia K, Luna J, Rust S, Huang Y. A natural language processing pipeline to synthesize patient-generated notes toward improving remote care and chronic disease management: a cystic fibrosis case study. JAMIA Open [Internet]. 2021;4(3). Available from: https://academic.oup.com/jamiaopen/article/doi/10.1093/jamiaopen/ooab084/6377936
https://doi.org/10.1093/jamiaopen/ooab084
 
57. Elyan E, Vuttipittayamongkol P, Johnston P, Martin K, McPherson K, Moreno-García CF, et al. Computer vision and machine learning for medical image analysis: recent advances, challenges, and way forward. Artif Intell Surg [Internet]. 2022; Available from: https://www.oaepublish.com/articles/ais.2021.15
https://doi.org/10.20517/ais.2021.15
 
58. Créquit P, Mansouri G, Benchoufi M, Vivot A, Ravaud P. Mapping of crowdsourcing in health: systematic review. J Med Internet Res [Internet]. 2018;20(5):e187. Available from: http://www.jmir.org/2018/5/e187/
https://doi.org/10.2196/jmir.9330
 
59. Yadav SS, Jadhav SM. Deep convolutional neural network based medical image classification for disease diagnosis. J Big Data [Internet]. 2019;6(1):113. Available from: https://journalofbigdata.springeropen.com/articles/10.1186/s40537-019-0276-2
https://doi.org/10.1186/s40537-019-0276-2
 
60. Dhaliwal A, Ma J, Zheng M, Lyu Q, Rajora MA, Ma S, et al. Deep learning for automatic organ and tumor segmentation in nanomedicine pharmacokinetics. Theranostics [Internet]. 2024;14(3):973-87. Available from: https://www.thno.org/v14p0973.htm
https://doi.org/10.7150/thno.90246
 
61. Arapi V, Hardt-Stremayr A, Weiss S, Steinbrener J. Bridging the simulation-to-real gap for AI-based needle and target detection in robot-assisted ultrasound-guided interventions. Eur Radiol Exp [Internet]. 2023;7(1):30. Available from: https://eurradiolexp.springeropen.com/articles/10.1186/s41747-023-00344-x
https://doi.org/10.1186/s41747-023-00344-x
 
62. Kitaguchi D, Takeshita N, Hasegawa H, Ito M. Artificial intelligence‐based computer vision in surgery: recent advances and future perspectives. Ann Gastroenterol Surg [Internet]. 2022;6(1):29-36. Available from: https://onlinelibrary.wiley.com/doi/10.1002/ags3.12513
https://doi.org/10.1002/ags3.12513
 
63. Bhaskar S, Bradley S, Sakhamuri S, Moguilner S, Chattu VK, Pandya S, et al. Designing futuristic telemedicine using artificial intelligence and robotics in the COVID-19 era. Front Public Heal [Internet]. 2020;8. Available from: https://www.frontiersin.org/articles/10.3389/fpubh.2020.556789/full
https://doi.org/10.3389/fpubh.2020.556789
 
64. Site A, Nurmi J, Lohan ES. Systematic review on machine-learning algorithms used in wearable-based eHealth data analysis. IEEE Access [Internet]. 2021;9:112221-35. Available from: https://ieeexplore.ieee.org/document/9508969/
https://doi.org/10.1109/ACCESS.2021.3103268
 
65. Manickam P, Mariappan SA, Murugesan SM, Hansda S, Kaushik A, Shinde R, et al. Artificial intelligence (AI) and internet of medical things (IoMT) assisted biomedical systems for intelligent healthcare. Biosensors [Internet]. 2022;12(8):562. Available from: https://www.mdpi.com/2079-6374/12/8/562
https://doi.org/10.3390/bios12080562
 
66. Reddy Kalli VD. Creating an AI-powered platform for neurosurgery alongside a usability examination: progressing towards minimally invasive robotics. J Artif Intell Gen Sci ISSN3006-4023 [Internet]. 2024;3(1):363-75. Available from: https://ojs.boulibrary.com/index.php/JAIGS/article/view/125
https://doi.org/10.60087/jaigs.v3i1.125
 
67. Tiwari R. Explainable AI (XAI) and its applications in building trust and understanding in AI decision making. Interantional J Sci Res Eng Manag [Internet]. 2023;07(01). Available from: https://ijsrem.com/download/explainable-ai-xai-and-its-applications-in- building-trust-and-understanding-in-ai-decision-making/
https://doi.org/10.55041/IJSREM17592
 
68. Rajkomar A, Dean J, Kohane I. Machine learning in medicine. N Engl J Med [Internet]. 2019;380(14):1347-58. Available from: http://www.nejm.org/doi/10.1056/NEJMra1814259
https://doi.org/10.1056/NEJMra1814259
 
69. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med [Internet]. 2019;25(1):44-56. Available from: https://www.nature.com/articles/s41591-018-0300-7
https://doi.org/10.1038/s41591-018-0300-7
 
70. Okumura M. Reply to Tian et al. Eur J Cardio-Thoracic Surg [Internet]. 2019; Available from: https://academic.oup.com/ejcts/advance- article/doi/10.1093/ejcts/ezz121/5475817
https://doi.org/10.1093/ejcts/ezz121
 
71. Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA [Internet]. 2016;316(22):2402. Available from: http://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.2016.17216
https://doi.org/10.1001/jama.2016.17216
 
72. Elhaddad M, Hamam S. AI-driven clinical decision support systems: an ongoing pursuit of potential. Cureus [Internet]. 2024; Available from: https://www.cureus.com/articles/244463-ai-driven-clinical-decision-support-systems-an- ongoing-pursuit-of-potential
https://doi.org/10.7759/cureus.57728
 
73. Reddy S, Allan S, Coghlan S, Cooper P. A governance model for the application of AI in health care. J Am Med Informatics Assoc [Internet]. 2020;27(3):491-7. Available from: https://academic.oup.com/jamia/article/27/3/491/5612169
https://doi.org/10.1093/jamia/ocz192
 
74. Massella M, Dri DA, Gramaglia D. Regulatory considerations on the use of machine learning based tools in clinical trials. Health Technol (Berl) [Internet]. 2022;12(6):1085-96. Available from: https://link.springer.com/10.1007/s12553-022-00708-0
https://doi.org/10.1007/s12553-022-00708-0
 
75. Bottomley D, Thaldar D. Liability for harm caused by AI in healthcare: an overview of the core legal concepts. Front Pharmacol [Internet]. 2023;14. Available from: https://www.frontiersin.org/articles/10.3389/fphar.2023.1297353/full
https://doi.org/10.3389/fphar.2023.1297353
 
76. Samek W, Müller KR. Towards explainable artificial intelligence. 2019. pp. 5-22. Available from: http://link.springer.com/10.1007/978-3-030-28954-6_1
https://doi.org/10.1007/978-3-030-28954-6_1

Cite this article as:
Fahmy Opportunities and Challenges of Applying Artificial Intelligence in Healthcare. ABC. Premier Journal of Artificial Intelligence 2024:1;100003

Export to EndNote Export to Mendeley