FDA-approved machine learning algorithms in neuroradiology: A systematic review of the current evidence for approval

Yearley, Alexander G.; Goedmakers, Caroline M W; Panahi, Armon; Rana, Aakanksha; Doucette, Joanne; Ranganathan, Kavitha; Smith, Timothy R.

doi:10.1016/j.artmed.2023.102607

Cited by 6 publications

(5 citation statements)

References 244 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent systematic review of 53 algorithms with applications in the central nervous system revealed a concerning lack of substantial peer review, with only 10 algorithms being published, and a scarcity of rigorous clinical efficacy studies, often involving biased study populations. 13 These findings underscore the importance of advocating for and actively participating in the creation of regulatory frameworks for the use of AI/ML in neurology and NM medicine. This is particularly important in preventing unforeseen and unintentional harms which may arise from the premature deployment of healthcare AI applications which may have had suboptimal development (including inadequate AI model training) and pre-market testing.…”

Section: Working With Industry and Regulatory Aspects Of Ai In Edx An...mentioning

confidence: 97%

Role of artificial intelligence in neuromuscular and electrodiagnostic medicine

2024

Muscle and Nerve

View full text Add to dashboard Cite

Section: Working With Industry and Regulatory Aspects Of Ai In Edx An...mentioning

confidence: 97%

Role of artificial intelligence in neuromuscular and electrodiagnostic medicine

2024

Muscle and Nerve

View full text Add to dashboard Cite

“…Another reason could be the high costs and resources associated with conducting rigorous clinical trials. The observed lack of a published research findings is not restricted to AI devices in mental healthcare but affects AI medical devices generally (Vokinger et al, 2021;Yearley et al, 2023).…”

Section: Clinical Evidencementioning

confidence: 99%

“…Regulatory oversight does not guarantee a product's external validity as the extent to which the results of a study can be generalized to and across other settings, individuals, and times (Andrade, 2018). Research has shown that AI tools have been approved with little or no efficacy data (Ebrahimian et al, 2022;van Leeuwen et al, 2021;Yearley et al, 2023). The absence of public information regarding the external validity of AI-CDSSs makes it difficult to justify their integration into clinical practice (Ebrahimian et al, 2022;Harada et al, 2021;Yearley et al, 2023).…”

mentioning

confidence: 99%

“…Research has shown that AI tools have been approved with little or no efficacy data (Ebrahimian et al, 2022;van Leeuwen et al, 2021;Yearley et al, 2023). The absence of public information regarding the external validity of AI-CDSSs makes it difficult to justify their integration into clinical practice (Ebrahimian et al, 2022;Harada et al, 2021;Yearley et al, 2023). Therefore, the second goal of the review is to provide information about available scientific evidence addressing the external validity of the identified tools to better understand their usefulness in patient care (Lesko et al, 2010).…”

mentioning

confidence: 99%

See 1 more Smart Citation

AI-Enabled Clinical Decision Support Tools for Mental Healthcare: A Product Review

Kleine,

Kokje,

Hummelsberger

et al. 2023

Preprint

View full text Add to dashboard Cite

This review seeks to promote transparency in the availability of regulated AI-enabled Clinical Decision Support Systems (AI-CDSSs) for mental healthcare. From an initial pool of 240 potential products, only seven fulfilled the inclusion criteria. The products can be categorized into three major areas: Autism Spectrum Disorder (ASD) diagnosis in children through behavioral data; multifaceted disorder diagnosis via conversational data; and depression medication prescription derived from clinical and genetic data. The exploration of unregulated products reveals the intricate challenges present in AI-CDSS regulation. Although we found five scientific articles evaluating the devices’ performance and external validity, the average completeness of reporting, indicated by a 52 percent adherence to the CONSORT-AI (Consolidated Standards of Reporting Trials Artificial Intelligence) checklist, was modest, signaling room for improvement in reporting quality.

show abstract

“…In recent years, with the increasing availability of medical data and the continuous improvement in computer analysis capabilities, machine learning has been increasingly used in the medical field [ 23 , 24 ]. Machine learning is a technological application that uses algorithms and data to enable computers to automatically learn and enhance.…”

Section: Introductionmentioning

confidence: 99%

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

Wang,

Chang,

Liu

et al. 2024

J Med Internet Res

View full text Add to dashboard Cite

Background Machine learning is a potentially effective method for predicting the response to platinum-based treatment for ovarian cancer. However, the predictive performance of various machine learning methods and variables is still a matter of controversy and debate. Objective This study aims to systematically review relevant literature on the predictive value of machine learning for platinum-based chemotherapy responses in patients with ovarian cancer. Methods Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we systematically searched the PubMed, Embase, Web of Science, and Cochrane databases for relevant studies on predictive models for platinum-based therapies for the treatment of ovarian cancer published before April 26, 2023. The Prediction Model Risk of Bias Assessment tool was used to evaluate the risk of bias in the included articles. Concordance index (C-index), sensitivity, and specificity were used to evaluate the performance of the prediction models to investigate the predictive value of machine learning for platinum chemotherapy responses in patients with ovarian cancer. Results A total of 1749 articles were examined, and 19 of them involving 39 models were eligible for this study. The most commonly used modeling methods were logistic regression (16/39, 41%), Extreme Gradient Boosting (4/39, 10%), and support vector machine (4/39, 10%). The training cohort reported C-index in 39 predictive models, with a pooled value of 0.806; the validation cohort reported C-index in 12 predictive models, with a pooled value of 0.831. Support vector machine performed well in both the training and validation cohorts, with a C-index of 0.942 and 0.879, respectively. The pooled sensitivity was 0.890, and the pooled specificity was 0.790 in the training cohort. Conclusions Machine learning can effectively predict how patients with ovarian cancer respond to platinum-based chemotherapy and may provide a reference for the development or updating of subsequent scoring systems.

show abstract

FDA-approved machine learning algorithms in neuroradiology: A systematic review of the current evidence for approval

Cited by 6 publications

References 244 publications

Role of artificial intelligence in neuromuscular and electrodiagnostic medicine

Role of artificial intelligence in neuromuscular and electrodiagnostic medicine

AI-Enabled Clinical Decision Support Tools for Mental Healthcare: A Product Review

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

Contact Info

Product

Resources

About