A Radiology-focused Review of Predictive Uncertainty for AI Interpretability in Computer-assisted Segmentation

McCrindle, Brian W.; Zukotynski, Katherine; Doyle, Thomas E.; Noseworthy, Michael D.

doi:10.1148/ryai.2021210031

Cited by 25 publications

(21 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…75 However, none of the studies identified through our review incorporated clinicians to systematically validate model explanations. Although there is a range of supporting literature, perspectives and reviews highlighting the need for interpretable machine learning in medical imaging, 11,27,29,76,77…”

Section: Remaining Challengesmentioning

confidence: 99%

Interpretable machine learning for dementia: A systematic review

Martin

Townend

Barkhof

et al. 2023

Alzheimer's & Dementia

View full text Add to dashboard Cite

Introduction:Machine learning research into automated dementia diagnosis is becoming increasingly popular but so far has had limited clinical impact. A key challenge is building robust and generalizable models that generate decisions that can be reliably explained. Some models are designed to be inherently "interpretable," whereas post hoc "explainability" methods can be used for other models.Methods: Here we sought to summarize the state-of-the-art of interpretable machine learning for dementia. Results:We identified 92 studies using PubMed, Web of Science, and Scopus. Studies demonstrate promising classification performance but vary in their validation procedures and reporting standards and rely heavily on popular data sets.Discussion: Future work should incorporate clinicians to validate explanation methods and make conclusive inferences about dementia-related disease pathology. Critically analyzing model explanations also requires an understanding of the interpretability methods itself. Patient-specific explanations are also required to demonstrate the benefit of interpretable machine learning in clinical practice.

show abstract

Section: Remaining Challengesmentioning

confidence: 99%

Interpretable machine learning for dementia: A systematic review

Martin

Townend

Barkhof

et al. 2023

Alzheimer's & Dementia

View full text Add to dashboard Cite

show abstract

“…The QUBIQ21 challenge aimed to quantify uncertainties in biomedical image segmentation. Recent advances in probabilistic deep learning allow for uncertainty estimation across predictions [69], which can pave the way to explainable, trustworthy AI and can inform clinicians about diagnostic uncertainty of AI [70]. QUBIQ21 addresses multiple organs and imaging modalities, including prostate MRI.…”

Section: Grand Challengesmentioning

confidence: 99%

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Sunoqrot

Saha²,

Hosseinzadeh³

et al. 2022

Eur Radiol Exp

View full text Add to dashboard Cite

Artificial intelligence (AI) for prostate magnetic resonance imaging (MRI) is starting to play a clinical role for prostate cancer (PCa) patients. AI-assisted reading is feasible, allowing workflow reduction. A total of 3,369 multi-vendor prostate MRI cases are available in open datasets, acquired from 2003 to 2021 in Europe or USA at 3 T (n = 3,018; 89.6%) or 1.5 T (n = 296; 8.8%), 346 cases scanned with endorectal coil (10.3%), 3,023 (89.7%) with phased-array surface coils; 412 collected for anatomical segmentation tasks, 3,096 for PCa detection/classification; for 2,240 cases lesions delineation is available and 56 cases have matching histopathologic images; for 2,620 cases the PSA level is provided; the total size of all open datasets amounts to approximately 253 GB. Of note, quality of annotations provided per dataset highly differ and attention must be paid when using these datasets (e.g., data overlap). Seven grand challenges and commercial applications from eleven vendors are here considered. Few small studies provided prospective validation. More work is needed, in particular validation on large-scale multi-institutional, well-curated public datasets to test general applicability. Moreover, AI needs to be explored for clinical stages other than detection/characterization (e.g., follow-up, prognosis, interventions, and focal treatment).

show abstract

“…In high volume datasets, a more quantitative metric that allows triaging of the images is preferred. 15 A probability score is an easily interpretable metric of the model confidence in assigning a J o u r n a l P r e -p r o o f 13 class. The model probability score was highly correlated with grader confidence and served as a useful metric to separate out mislabeled images, especially for good quality images.…”

Section: J O U R N a L P R E -P R O O Fmentioning

confidence: 99%