Abstract
Artificial intelligence (AI) has shown transformative potential in various medical fields, including diagnostic imaging. Recent advances in AI-driven technologies have opened new avenues for improving echocardiographic practices. AI algorithms enhance the image quality, automate measurements, and assist in the diagnosis of cardiovascular diseases. These technologies reduce manual errors, increase consistency, and match the diagnostic performances of experienced echocardiographers. AI in tele-echocardiography offers significant benefits, particularly in rural and remote regions in Japan, where healthcare provider shortages and geographic isolation hinder access to advanced medical care. AI enhances accessibility, provides real-time remote analyses, supports continuous monitoring, and improves the quality and efficiency of remotely delivered cardiac care. However, addressing challenges related to data security, transparency, integration into clinical workflows, and ethical considerations is essential for the successful implementation of AI in echocardiography. On overcoming these challenges, AI will be able to revolutionize echocardiography and ensure timely and effective cardiac care for all patients in the future.
Keywords: artificial intelligence, echocardiography, diagnostic accuracy, workflow optimization, patient care, deep learning
Introduction
Echocardiography has evolved significantly since its start, with over several million echocardiograms being performed annually worldwide (1,2). It remains a critical diagnostic tool in cardiology. Artificial intelligence (AI), particularly machine and deep learning, has shown transformative potential across various medical fields, including diagnostic imaging (3,4).
Recent advancements in AI-driven technologies have opened new avenues for improving echocardiographic practices. AI algorithms can enhance the image quality (5,6), automate measurements (7,8), and assist in the diagnosis of cardiovascular diseases (9,10). For example, AI has enabled significant improvements in image analyses, with deep learning algorithms achieving accuracy rates of up to 98% for recognizing and classifying cardiac structures (11). These technologies also automate measurements of ventricular volumes, ejection fraction, and other critical parameters, thereby reducing manual errors and increasing consistency (12,13). Numerous studies have demonstrated the clinical benefits of AI in echocardiography, such as reduced inter-operator variability and matching the diagnostic performance of experienced echocardiographers.
Despite these promising advancements, however, several challenges remain to be overcome. One major challenge is the variability in the AI algorithm performance across different patient populations and clinical settings (14). Extensive validation studies are required to ensure the reliability and effectiveness of these tools in many situations. The quality of echocardiographic images can significantly impact the performance of AI algorithms, standardize imaging protocols, and ensure high-quality data input that is critical for optimizing AI applications (15). Ethical concerns related to data privacy and the potential of AI to replace human jobs in healthcare need to be addressed (16).
This review aims to provide a comprehensive overview of the current and future applications of AI in echocardiography. By synthesizing findings from recent studies and clinical trials, we seek to highlight the benefits and challenges of integrating AI into echo-labs.
AI in Echocardiography: Current Landscape (Fig. 1)
Figure 1.
Current AI Landscape in Echocardiography. This figure illustrates the current advancements and applications of artificial intelligence (AI) in echocardiography. It highlights four key areas where AI has made significant contributions.
Echocardiography has long been the cornerstone of cardiovascular diagnostics, providing critical insights into the cardiac structure and function. The integration of AI in echocardiography can significantly enhance the diagnostic accuracy and efficiency. This section explores the current landscape of AI applications in echocardiography, detailing advancements, benefits, and potential challenges. Several key advancements are detailed below (Table 1).
Table 1.
AI Advancements in Echocardiography.
| Application | Description | Benefits | Examples |
|---|---|---|---|
| Image Optimization and Acquisition | AI algorithms optimize ultrasound images, reducing scanning time and eliminating artifacts. | Enhances image clarity, ensures consistency, improves diagnostic accuracy. | AI algorithms are used to enhance blurry images, integrated into commercial machines. |
| Image Segmentation and Measurement | AI accurately delineates cardiac structures, such as chambers and valves, and automates. measurements of key parameters. | Reduces clinician workload, minimizes human error, provides consistent and reliable data. | AI trained to classify echocardiographic views. |
| Disease Detection and Classification | AI detects abnormalities and classifies cardiac conditions. | Matches or exceeds diagnostic performance of experienced cardiologists, supports clinical decisions. | AI identifies patterns associated with specific cardiovascular diseases. |
| Workflow Optimization | AI automates routine tasks such as image acquisition, assessment, and report generation. | Improves patient throughput, reduces time to diagnosis, allows focus on complex decision-making. | AI systems streamline workflow in echo labs, facilitating efficient follow-up and monitoring. |
AI: artificial intelligence
1) Image optimization and acquisition
Deep-learning algorithms are employed to enhance image clarity, particularly in blurry or suboptimal images (17). This technology has been integrated into commercially available ultrasound machines, making advanced image enhancement accessible in routine clinical practice (18).
2) Image segmentation and measurement
One of the significant applications of AI in echocardiography is the segmentation and measurement of the cardiac structures (19,20). AI algorithms can accurately delineate the borders of the cardiac chambers, valves, and other structures (21). This is also shown by our study, where a convolutional neural network was trained to classify echocardiographic views with impressive accuracy (11). This precision is crucial for obtaining measurements of ventricular volume, ejection fraction, and other critical parameters (22).
3) Disease detection and classification
AI has made significant strides in disease detection and classification (23-25). Deep learning algorithms can analyze echocardiographic images to detect abnormalities and classify various cardiac conditions. For example, AI can identify patterns associated with diseases, such as hypertrophic cardiomyopathy (26,27), amyloidosis (28,29), and pulmonary hypertension (9,30). These algorithms can match or even exceed the diagnostic performance of experienced cardiologists and provide valuable support for clinical decision-making (31).
4) Workflow optimization
The integration of AI into echocardiography streamlines the workflow in echo-labs. AI systems can automate routine tasks, such as image acquisition, assessments, and report generation (32). This automation allows healthcare professionals to focus on interpreting results and making clinical decisions, ultimately improving patient throughput and reducing the time to make a diagnosis (12).
Benefits of AI Integration
The integration of AI into echocardiography offers numerous benefits, including workflow optimization, enhanced diagnostic accuracy, and improved patient care. By automating routine tasks, such as image acquisition and measurement calculations, AI allows healthcare professionals to focus on more complex cases and decision-making processes. This optimization leads to faster echocardiography reporting times and improved patient throughput in healthcare facilities.
1) Enhanced diagnostic accuracy
AI-driven algorithms can significantly enhance echocardiography diagnostic accuracy. By automating the measurement of ventricular volumes, ejection fraction, and other critical parameters, AI reduces human error and ensures consistent and reliable data. In addition, AI can detect and classify various cardiac conditions with high accuracy, providing valuable support for clinical decision-making (33).
2) Improved patient care
AI integration into echocardiography improves patient care by providing more accurate and timely diagnoses (34). Faster reporting and enhanced diagnostic accuracy enable healthcare professionals to make quicker and more informed clinical decisions, leading to better patient outcomes. In addition, AI can help identify subtle changes in the cardiac function that may be missed by human observers, allowing for earlier detection and intervention in cardiovascular diseases (35).
3) Cost efficiency
AI enhances the cost efficiency of echocardiography by reducing the operational costs and improving financial sustainability (36). Automating routine tasks reduces the need for manual labor, minimizes errors, and improves the overall efficiency. While the initial implementation of AI systems may involve significant costs, long-term savings from increased efficiency and reduced errors can outweigh these expenses (37).
Benefits of AI in Tele-echocardiography
In Japan, there are significant healthcare challenges in rural and remote regions, including sparsely populated areas and isolated islands (e.g., Okinawa Prefecture). These regions often experience a shortage of healthcare providers, particularly specialists, which limits access to advanced medical care (38). The integration of AI into tele-echocardiography presents a promising solution to these challenges by enhancing the accessibility and quality of cardiac care (39).
According to previous reports, the number of physicians per capita in rural regions is significantly lower than that in urban centers. This disparity leads to long wait times for medical consultations and difficulty in accessing specialized care, such as cardiology. Japan's geographic features include numerous isolated islands and mountainous regions where transportation can be challenging. Residents in these areas often must travel long distances to access healthcare facilities, which can be particularly burdensome for elderly patients and those with chronic conditions. The limited transportation infrastructure further exacerbates the difficulty in accessing timely medical care. In addition, Japan's population is aging rapidly, with rural areas experiencing higher rates of aging than urban centers (40). The elderly population often requires more frequent and specialized medical care, including cardiac services. The lack of local specialized care facilities means that elderly patients face significant challenges in effectively managing their health conditions.
The integration of AI in tele-echocardiography offers several key benefits, significantly enhancing the accessibility, quality, and efficiency of remotely delivered cardiac care (Fig. 2). These benefits are particularly impactful in extending high-quality health care services to underserved and rural areas.
Figure 2.
Tele-echocardiography. This figure illustrates the use of AI in tele-echocardiography, showcasing its role in providing remote, real-time cardiac diagnostics, enhancing accessibility to high-quality cardiac care in rural and remote areas.
1) Enhanced accessibility
AI-powered tele-echocardiography systems enable healthcare providers to extend their services to remote and rural areas, where access to specialized cardiac care is limited. By utilizing AI for remote diagnostics, patients can receive high-quality echocardiographic evaluations without the need for travel to specialized centers. This improvement in accessibility ensures that patients in geographically isolated areas receive timely and necessary cardiac care, thereby reducing disparities in healthcare delivery (41).
2) Real-time remote analyses
AI algorithms facilitate the real-time analysis of echocardiographic images, providing immediate feedback and preliminary diagnoses (42). This capability is crucial in emergency situations and when immediate clinical decisions are required. Remote healthcare providers can leverage AI's diagnostic capabilities to promptly identify critical conditions, such as acute myocardial infarction or severe valvular disease, thereby improving patient outcomes through timely interventions.
3) Automated image acquisition and quality control
AI systems assist nonspecialist operators in acquiring high-quality echocardiographic images. These systems provide real-time guidance and automatically adjust the imaging parameters to ensure that the captured images are diagnostically useful. This automation reduces the variability in the image quality that can result from disparity in operator skill levels, thereby ensuring consistent and reliable diagnostics across various settings (43).
4) Cost-effective healthcare delivery
The use of AI in tele-echocardiography reduces the need for frequent specialist visits and the associated travel costs (44), by automating many aspects of image acquisition and analyses.
5) Continuous monitoring and follow-up
AI-enabled tele-echocardiography systems support the continuous monitoring and follow-up of patients with chronic cardiac conditions. Remote monitoring facilitated by AI allows for early detection of disease progression or complications, enabling timely interventions that can prevent hospital readmissions and improve long-term patient outcomes (45).
Potential Challenges and Ethical Considerations
Although AI offers many benefits in echocardiography, its integration presents several challenges and ethical considerations that need to be addressed to ensure successful implementation (Table 2).
Table 2.
Potential Challenges and Ethical Considerations in AI Integration for Echocardiography.
| Challenge | Description | Impact |
|---|---|---|
| Data Security and Privacy | Safeguarding patient data by ensuring robust security measures such as encryption, secure storage, and access control. | Maintains regulatory compliance, protects patient privacy, builds trust in AI systems. |
| Transparency and Explainability | Addressing the “black box” nature of AI by making algorithms transparent and their outputs explainable to clinicians. | Enhances trust in AI, supports clinical decision-making, ensures AI’s decisions are clear and understandable. |
| Integration with Clinical Workflow | Seamlessly incorporating AI tools into existing workflows without significant disruptions. | Ensures smooth transition, effective utilization of AI, maintains workflow efficiency. |
| Ethical Concerns | Addressing issues such as bias in AI algorithms, potential job displacement, and impact on clinician-patient relationships. | Maintains public trust, ensures ethical use of AI, maximizes benefits of AI while minimizing negative impacts. |
AI: artificial intelligence
1) Data security and privacy
One of the foremost concerns is safeguarding patient data. AI systems handle and store sensitive patient information, thus making robust security measures essential for maintaining regulatory compliance and protecting patient privacy. Strict adherence to regulations is crucial (46).
2) Transparency and explainability
The “black box” nature of some AI algorithms poses a challenge for clinical decision-making (47). Understanding and interpreting the outputs of complex AI models can be difficult, thus necessitating transparency and explainability in AI applications. Clinicians must trust the AI systems they use, which requires these systems to provide clear and understandable reasons for their decisions.
3) Integration with clinical workflow
Integrating AI systems into existing clinical workflows is challenging (48). AI tools must be seamlessly incorporated into the daily practices of healthcare professionals without causing significant disruptions. This requires careful planning, training, and redesigning of the workflow to accommodate new technologies. To effectively utilize AI in echocardiography, healthcare professionals require adequate training and education. Understanding how AI systems work, their limitations, and how to interpret their output is crucial for clinicians.
4) Ethical concerns
Ethical considerations are essential for integrating AI into echocardiography. Issues such as bias in AI algorithms, the potential for AI to replace human jobs, and the impact on the clinician-patient relationship must be addressed (49). Ensuring AI systems are designed and used ethically is essential for maintaining public trust and maximizing the benefits of AI.
Conclusion
AI has the potential to revolutionize echocardiography by offering significant improvements in diagnostic accuracy, workflow efficiency, patient care, and tele-echocardiography. As AI technology continues to advance, its integration into routine clinical practice is becoming increasingly indispensable. However, addressing challenges related to data security, ethical considerations, and the need for human-AI collaboration is essential for the successful implementation of AI in echocardiography (50).
The author states that he has no Conflict of Interest (COI).
Financial Support
This work was partially supported by grants from JSPS Kakenhi Grants (Number 23K07509 to K. Kusunose) and the Japan Agency for Medical Research and Development (AMED, JP22uk1024007 to K.K.).
References
- 1.Vaid A, Duong SQ, Lampert J, et al. Local large language models for privacy-preserving accelerated review of historic echocardiogram reports. J Am Med Inform Assoc. Forthcoming. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Nedadur R, Wang B, Tsang W. Artificial intelligence for the echocardiographic assessment of valvular heart disease. Heart 108: 1592-1599, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kusunose K. Steps to use artificial intelligence in echocardiography. J Echocardiogr 19: 21-27, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Akkus Z, Aly YH, Attia IZ, et al. Artificial intelligence (AI)-empowered echocardiography interpretation: a state-of-the-art review. J Clin Med 10: 1391, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Schuuring MJ, Išgum I, Cosyns B, Chamuleau SA, Bouma BJ. Routine echocardiography and artificial intelligence solutions. Front Cardiovasc Med 8: 648877, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Huang KC, Huang CS, Su MY, et al. Artificial intelligence aids cardiac image quality assessment for improving precision in strain measurements. JACC Cardiovasc Imaging 14: 335-345, 2021. [DOI] [PubMed] [Google Scholar]
- 7.Karužas A, Balčiūnas J, Fukson M, et al. Artificial intelligence for automated evaluation of aortic measurements in 2D echocardiography: feasibility, accuracy, and reproducibility. Echocardiography 39: 1439-1445, 2022. [DOI] [PubMed] [Google Scholar]
- 8.Kusunose K, Zheng R, Yamada H, Sata M. How to standardize the measurement of left ventricular ejection fraction. J Med Ultrasonics 49: 35-43, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hirata Y, Tsuji T, Kotoku Ji, Sata M, Kusunose K. Echocardiographic artificial intelligence for pulmonary hypertension classification. Heart 110: 586-593, 2024. [DOI] [PubMed] [Google Scholar]
- 10.Krishna H, Desai K, Slostad B, et al. Fully automated artificial intelligence assessment of aortic stenosis by echocardiography. J Am Soc Echocardiogr 36: 769-777, 2023. [DOI] [PubMed] [Google Scholar]
- 11.Kusunose K, Haga A, Inoue M, Fukuda D, Yamada H, Sata M. Clinically feasible and accurate view classification of echocardiographic images using deep learning. Biomolecules 10: 665, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hirata Y, Nomura Y, Saijo Y, Sata M, Kusunose K. Reducing echocardiographic examination time through routine use of fully automated software: a comparative study of measurement and report creation time. J Echocardiogr. Forthcoming. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yamaguchi N, Kosaka Y, Haga A, Sata M, Kusunose K. Artificial intelligence-assisted interpretation of systolic function by echocardiogram. Open Heart 10: e002287, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Alowais SA, Alghamdi SS, Alsuhebany N, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ 23: 689, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xu J, Wu B, Huang J, Gong Y, Zhang Y, Liu B. Practical applications of advanced cloud services and generative AI systems in medical image analysis. arXiv: 2403.17549, 2024. [Google Scholar]
- 16.Fülöp MT, Topor DI, Ionescu CA, Cifuentes-Faura J, Măgdaş N. Ethical concerns associated with artificial intelligence in the accounting profession: a curse or a blessing? J Bus Econ Manage 24: 387-404, 2023. [Google Scholar]
- 17.Stewart JE, Goudie A, Mukherjee A, Dwivedi G. Artificial intelligence-enhanced echocardiography in the emergency department. Emerg Med Australa 33: 1117-1120, 2021. [DOI] [PubMed] [Google Scholar]
- 18.Taksoee-Vester CA, Mikolaj K, Bashir Z, et al. AI supported fetal echocardiography with quality assessment. Sci Rep 14: 5809, 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shi F, Wang J, Shi J, et al. Review of artificial intelligence techniques in imaging data acquisition, segmentation, and diagnosis for COVID-19. IEEE Rev Biomed Eng 14: 4-15, 2020. [DOI] [PubMed] [Google Scholar]
- 20.Wegner FK, Benesch Vidal ML, Niehues P, et al. Accuracy of deep learning echocardiographic view classification in patients with congenital or structural heart disease: importance of specific datasets. J Clin Med 11: 690, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Amer A, Ye X, Janan F. ResDUnet: a deep learning-based left ventricle segmentation method for echocardiography. IEEE Access 9: 159755-159763, 2021. [Google Scholar]
- 22.Kusunose K, Haga A, Yamaguchi N, et al. Deep learning for assessment of left ventricular ejection fraction from echocardiographic images. J Am Soc Echocardiogr 33: 632-635.e1, 2020. [DOI] [PubMed] [Google Scholar]
- 23.Wahlang I, Maji AK, Saha G, et al. Deep Learning methods for classification of certain abnormalities in echocardiography. Electronics 10: 495, 2021. [Google Scholar]
- 24.Pandey A, Kagiyama N, Yanamala N, et al. Deep-learning models for the echocardiographic assessment of diastolic dysfunction. Cardiovasc Imaging 14: 1887-1900, 2021. [DOI] [PubMed] [Google Scholar]
- 25.Kusunose K, Abe T, Haga A, et al. A deep learning approach for assessment of regional wall motion abnormality from echocardiographic images. Cardiovasc Imaging 13: 374-381, 2020. [DOI] [PubMed] [Google Scholar]
- 26.Morita SX, Kusunose K, Haga A, et al. Deep learning analysis of echocardiographic images to predict positive genotype in patients with hypertrophic cardiomyopathy. Front Cardiovasc Med 8: 669860, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Farahani NZ, Enayati M, Sundaram DSB, et al. Application of machine learning for detection of hypertrophic cardiomyopathy patients from echocardiogram measurements. Proceedings of the 2021 Design of Medical Devices Conference; 2021 Apr 12-15; Minneapolis, USA. Amsterdam, Elsevier, 2021. [Google Scholar]
- 28.Zhang X, Liang T, Su C, et al. Deep learn-based computer-assisted transthoracic echocardiography: approach to the diagnosis of cardiac amyloidosis. Int J Cardiovasc Imaging 39: 955-965, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Boldrini M, Cappelli F, Chacko L, et al. Multiparametric echocardiography scores for the diagnosis of cardiac amyloidosis. Cardiovasc Imaging 13: 909-920, 2020. [DOI] [PubMed] [Google Scholar]
- 30.Murayama M, Sugimori H, Yoshimura T, et al. Deep learning to assess right ventricular ejection fraction from two-dimensional echocardiograms in precapillary pulmonary hypertension. Echocardiography 41: e15812, 2024. [DOI] [PubMed] [Google Scholar]
- 31.Yu X, Yao X, Wu B, et al. Using deep learning method to identify left ventricular hypertrophy on echocardiography. Int J Cardiovasc Imaging 38: 759-769, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kusunose K. Revolution of echocardiographic reporting: the new era of artificial intelligence and natural language processing. J Echocardiogr 21: 99-104, 2023. [DOI] [PubMed] [Google Scholar]
- 33.Venigandla K. Integrating RPA with AI and ML for enhanced diagnostic accuracy in healthcare. Power Syst Technol 46: 2022. [Google Scholar]
- 34.Ben-Israel D, Jacobs WB, Casha S, et al. The impact of machine learning on patient care: a systematic review. Artif Intell Med 103: 101785, 2020. [DOI] [PubMed] [Google Scholar]
- 35.Fouad H, Hassanein AS, Soliman AM, Al-Feel H. Analyzing patient health information based on IoT sensor with AI for improving patient assistance in the future direction. Measurement 159: 107757, 2020. [Google Scholar]
- 36.Gahungu N, Trueick R, Bhat S, Sengupta PP, Dwivedi G. Current challenges and recent updates in artificial intelligence and echocardiography. Cur Cardiovasc Imaging Rep 13: 5, 2020. [Google Scholar]
- 37.Do J, Ferreira VC, Bobarshad H, et al. Cost-effective, energy-efficient, and scalable storage computing for large-scale AI applications. ACM Trans Storage (TOS) 16: 1-37, 2020. [Google Scholar]
- 38.Karako K, Song P, Chen Y, Tang W, Kokudo N. Overview of the characteristics of and responses to the three waves of COVID-19 in Japan during 2020-2021. Biosci Trends 15: 1-8, 2021. [DOI] [PubMed] [Google Scholar]
- 39.Bolin EH, Collins RT 2nd, Best TH, et al. Implementation of a statewide, multisite fetal tele-echocardiography program: evaluation of more than 1100 fetuses over 9 years. J Perinatol 40: 1524-1530, 2020. [DOI] [PubMed] [Google Scholar]
- 40.Tamiya N, Noguchi H, Nishi A, et al. Population ageing and wellbeing: lessons from Japan's long-term care insurance policy. Lancet 378: 1183-1192, 2011. [DOI] [PubMed] [Google Scholar]
- 41.Shibuya K, Hashimoto H, Ikegami N, et al. Future of Japan's system of good health at low cost with equity: beyond universal coverage. Lancet 378: 1265-1273, 2011. [DOI] [PubMed] [Google Scholar]
- 42.Kakria P, Tripathi NK, Kitipawang P. A real-time health monitoring system for remote cardiac patients using smartphone and wearable sensors. Int J Telemed Appl 2015: 373474, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Armstrong AC, Ricketts EP, Cox C, et al. Quality control and reproducibility in M-mode, two-dimensional, and speckle tracking echocardiography acquisition and analysis: the CARDIA study, year 25 examination experience. Echocardiography 32: 1233-1240, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lewin S, Lavis JN, Oxman AD, et al. Supporting the delivery of cost-effective interventions in primary health-care systems in low-income and middle-income countries: an overview of systematic reviews. Lancet 372: 928-939, 2008. [DOI] [PubMed] [Google Scholar]
- 45.Bernstein RA, Kamel H, Granger CB, et al.; the STROKE-AF Investigators . Effect of long-term continuous cardiac monitoring vs usual care on detection of atrial fibrillation in patients with stroke attributed to large- or small-vessel disease: the STROKE-AF randomized clinical trial. JAMA 325: 2169-2177, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Mrabet H, Belguith S, Alhomoud A, Jemai A. A survey of IoT security based on a layered architecture of sensing and data analysis. Sensors (Basel) 20: 3625, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Poon AIF, Sung JJY. Opening the black box of AI-Medicine. J Gastroenterol Hepatol 36: 581-584, 2021. [DOI] [PubMed] [Google Scholar]
- 48.Juluru K, Shih HH, Keshava Murthy KN, et al. Integrating Al algorithms into the clinical workflow. Radiol Artif Intell 3: e210013, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Tippins NT, Oswald FL, McPhail SM. Scientific, legal, and ethical concerns about AI-based personnel selection tools: a call to action. Personnel Assessment and Decisions 7: 1-22, 2021. [Google Scholar]
- 50.Reverberi C, Rigon T, Solari A, et al.; GI Genius CADx Study Group . Experimental evidence of effective human-AI collaboration in medical decision-making. Sci Rep 12: 14952, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]