The need for uncertainty quantification in machine-assisted medical decision making

Begoli, Edmon; Bhattacharya, Tanmoy; Kusnezov, Dimitri F.

doi:10.1038/s42256-018-0004-1

The need for uncertainty quantification in machine-assisted medical decision making

Journal Article · Sun Jan 06 23:00:00 EST 2019 · Nature Machine Intelligence

DOI:https://doi.org/10.1038/s42256-018-0004-1· OSTI ID:1561669

^[1]; Bhattacharya, Tanmoy ^[2]; Kusnezov, Dimitri F. ^[3]

Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Dept. of Energy (DOE), Washington DC (United States)

We present that medicine, even from the earliest days of artificial intelligence (AI) research, has been one of the most inspiring and promising domains for the application of AI-based approaches. Equally, it has been one of the more challenging areas to see an effective adoption. There are many reasons for this, primarily the reluctance to delegate decision making to machine intelligence in cases where patient safety is at stake. To address some of these challenges, medical AI, especially in its modern data-rich deep learning guise, needs to develop a principled and formal uncertainty quantification (UQ) discipline, just as we have seen in fields such as nuclear stockpile stewardship and risk management. The data-rich world of AI-based learning and the frequent absence of a well-understood underlying theory poses its own unique challenges to straightforward adoption of UQ. These challenges, while not trivial, also present significant new research opportunities for the development of new theoretical approaches, and for the practical applications of UQ in the area of machine-assisted medical decision making. Finally, understanding prediction system structure and defensibly quantifying uncertainty is possible, and, if done, can significantly benefit both research and practical applications of AI in this critical domain.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC05-00OR22725; 89233218CNA000001

OSTI ID:: 1561669

Journal Information:: Nature Machine Intelligence, Journal Name: Nature Machine Intelligence Journal Issue: 1 Vol. 1; ISSN 2522-5839

Publisher:: The Author(s), Springer NatureCopyright Statement

Country of Publication:: United States

Language:: English

References (21)

Cancer Moonshot Data and Technology Team: Enabling a National Learning Healthcare System for Cancer to Unleash the Power of Data Hsu, Er; Klemm, Jd; Kerlavage, Ar Clinical Pharmacology & Therapeutics, Vol. 101, Issue 5 https://doi.org/10.1002/cpt.636	journal	April 2017
MR image-based radiomics to differentiate type Ι and type ΙΙ epithelial ovarian cancers Jian, Junming; Li, Yong’ai; Pickhardt, Perry J. European Radiology, Vol. 31, Issue 1 https://doi.org/10.1007/s00330-020-07091-2	journal	August 2020
Automatic discrimination of human hematopoietic tumor cell lines using a combination of imaging flow cytometry and convolutional neural network Matsuoka, Yoshikazu; Nakatsuka, Ryusuke; Fujioka, Tatsuya Human Cell, Vol. 34, Issue 3 https://doi.org/10.1007/s13577-021-00506-2	journal	February 2021
Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning Kermany, Daniel S.; Goldbaum, Michael; Cai, Wenjia Cell, Vol. 172, Issue 5 https://doi.org/10.1016/j.cell.2018.02.010	journal	February 2018
Prediction of Organic Reaction Outcomes Using Machine Learning Coley, Connor W.; Barzilay, Regina; Jaakkola, Tommi S. ACS Central Science, Vol. 3, Issue 5 https://doi.org/10.1021/acscentsci.7b00064	journal	April 2017
Dermatologist-level classification of skin cancer with deep neural networks Esteva, Andre; Kuprel, Brett; Novoa, Roberto A. Nature, Vol. 542, Issue 7639 https://doi.org/10.1038/nature21056	journal	January 2017
Comparison of methods that use whole genome data to estimate the heritability and genetic architecture of complex traits Evans, Luke M.; Tahmasbi, Rasool; Vrieze, Scott I. Nature Genetics, Vol. 50, Issue 5 https://doi.org/10.1038/s41588-018-0108-x	journal	April 2018
Deep learning based tissue analysis predicts outcome in colorectal cancer Bychkov, Dmitrii; Linder, Nina; Turkki, Riku Scientific Reports, Vol. 8, Issue 1 https://doi.org/10.1038/s41598-018-21758-3	journal	February 2018
Scalable and accurate deep learning with electronic health records Rajkomar, Alvin; Oren, Eyal; Chen, Kai npj Digital Medicine, Vol. 1, Issue 1 https://doi.org/10.1038/s41746-018-0029-1	journal	May 2018
The mystery of missing heritability: Genetic interactions create phantom heritability Zuk, O.; Hechter, E.; Sunyaev, S. R. Proceedings of the National Academy of Sciences, Vol. 109, Issue 4 https://doi.org/10.1073/pnas.1119675109	journal	January 2012
Artificial intelligence for melanoma diagnosis: how can we deliver on the promise? Mar, V. J.; Soyer, H. P. Annals of Oncology, Vol. 30, Issue 12 https://doi.org/10.1093/annonc/mdy191	journal	December 2019
Artificial intelligence for melanoma diagnosis: how can we deliver on the promise? Mar, V. J.; Soyer, H. P. Annals of Oncology, Vol. 29, Issue 8 https://doi.org/10.1093/annonc/mdy193	journal	May 2018
Why Deep Learning Works: A Manifold Disentanglement Perspective Brahma, Pratik Prabhanjan; Wu, Dapeng; She, Yiyuan IEEE Transactions on Neural Networks and Learning Systems, Vol. 27, Issue 10 https://doi.org/10.1109/TNNLS.2015.2496947	journal	October 2016
Applying deep neural networks to unstructured text notes in electronic medical records for phenotyping youth depression Geraci, Joseph; Wilansky, Pamela; de Luca, Vincenzo Evidence Based Mental Health, Vol. 20, Issue 3 https://doi.org/10.1136/eb-2017-102688	journal	July 2017
Can machine-learning improve cardiovascular risk prediction using routine clinical data? Weng, Stephen F.; Reps, Jenna; Kai, Joe PLOS ONE, Vol. 12, Issue 4 https://doi.org/10.1371/journal.pone.0174944	journal	April 2017
Readmission prediction via deep contextual embedding of clinical concepts Xiao, Cao; Ma, Tengfei; Dieng, Adji B. PLOS ONE, Vol. 13, Issue 4 https://doi.org/10.1371/journal.pone.0195024	journal	April 2018
Open Category Classification by Adversarial Sample Generation Yu, Yang; Qu, Wei-Yang; Li, Nan Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence https://doi.org/10.24963/ijcai.2017/469	conference	August 2017
Generative OpenMax for Multi-Class Open Set Classification Ge, ZongYuan; Demyanov, Sergey; Chen, Zetao arXiv https://doi.org/10.48550/arxiv.1707.07418	preprint	January 2017
Generative OpenMax for Multi-Class Open Set Classification Ge, Zongyuan; Demyanov, Sergey; Garnavi, Rahil Procedings of the British Machine Vision Conference 2017 https://doi.org/10.5244/C.31.42	conference	January 2017
Interplay between Epigenetics and Genetics in Cancer Choi, Jae Duk; Lee, Jong-Soo Genomics & Informatics, Vol. 11, Issue 4 https://doi.org/10.5808/GI.2013.11.4.164	journal	January 2013
Readmission prediction via deep contextual embedding of clinical concepts Xiao, Cao; Ma, Tengfei; Dieng, Adji Bousso Columbia University https://doi.org/10.7916/d82v3zbg	text	January 2018

Cited By (14)

Designing Human-Agent Collaborations: Commitment, responsiveness, and support Cila, Nazli CHI Conference on Human Factors in Computing Systems https://doi.org/10.1145/3491102.3517500	conference	April 2022
Ethical considerations about artificial intelligence for prognostication in intensive care Beil, Michael; Proft, Ingo; van Heerden, Daniel Intensive Care Medicine Experimental, Vol. 7, Issue 1 https://doi.org/10.1186/s40635-019-0286-6	journal	December 2019
Risk of estimators for Sobol’ sensitivity indices based on metamodels Panin, Ivan Electronic Journal of Statistics, Vol. 15, Issue 1 https://doi.org/10.1214/20-ejs1793	journal	January 2021
Demystifying Brain Tumor Segmentation Networks: Interpretability and Uncertainty Analysis Natekar, Parth; Kori, Avinash; Krishnamurthi, Ganapathy Frontiers in Computational Neuroscience, Vol. 14 https://doi.org/10.3389/fncom.2020.00006	journal	February 2020
AI Meets Exascale Computing: Advancing Cancer Research With Large-Scale High Performance Computing Bhattacharya, Tanmoy; Brettin, Thomas; Doroshow, James H. Frontiers in Oncology, Vol. 9 https://doi.org/10.3389/fonc.2019.00984	journal	October 2019
Uncertainty Bounds for Multivariate Machine Learning Predictions on High-Strain Brittle Fracture Garcia-Cardona, Cristina; Fernández-Godino, M. Giselle; O'Malley, Daniel arXiv https://doi.org/10.48550/arxiv.2012.15739	preprint	January 2020
EUCA: the End-User-Centered Explainable AI Framework Jin, Weina; Fan, Jianyu; Gromala, Diane arXiv https://doi.org/10.48550/arxiv.2102.02437	preprint	January 2021
Interpretable multimodal deep learning for real-time pan-tissue pan-disease pathology search on social media Schaumberg, Andrew J.; Juarez-Nicanor, Wendy C.; Choudhury, Sarah J. Modern Pathology, Vol. 33, Issue 11 https://doi.org/10.1038/s41379-020-0540-1	journal	May 2020
The Bionic Radiologist: avoiding blurry pictures and providing greater insights Dewey, Marc; Wilkens, Uta npj Digital Medicine, Vol. 2, Issue 1 https://doi.org/10.1038/s41746-019-0142-9	journal	July 2019
A Bayesian graph convolutional network for reliable prediction of molecular properties with uncertainty quantification Ryu, Seongok; Kwon, Yongchan; Kim, Woo Youn Chemical Science, Vol. 10, Issue 36 https://doi.org/10.1039/c9sc01992h	journal	January 2019
Augmented decision‐making for acral lentiginous melanoma detection using deep convolutional neural networks Lee, S.; Chu, Y. S.; Yoo, S. K. Journal of the European Academy of Dermatology and Venereology, Vol. 34, Issue 8 https://doi.org/10.1111/jdv.16185	journal	January 2020
Responsible AI: requirements and challenges Ghallab, Malik AI Perspectives, Vol. 1, Issue 1 https://doi.org/10.1186/s42467-019-0003-z	journal	September 2019
The Bionic Radiologist: avoiding blurry pictures and providing greater insights Dewey, Marc; Wilkens, Uta Charité - Universitätsmedizin Berlin https://doi.org/10.17169/refubium-4122	text	January 2019
Unlocking the Power of Artificial Intelligence and Big Data in Medicine Lovis, Christian Journal of Medical Internet Research, Vol. 21, Issue 11 https://doi.org/10.2196/16607	journal	January 2019

Similar Records

AI@DOE Interim Executive Report

Technical Report · Tue Nov 01 00:00:00 EDT 2022 · OSTI ID:1872103

Lowering the computational barrier: Partially Bayesian neural networks for transparency in medical imaging AI

Journal Article · Tue Feb 14 23:00:00 EST 2023 · Frontiers in Computer Science · OSTI ID:1924814

Related Subjects

97 MATHEMATICS AND COMPUTING
Computer science
information technology

The need for uncertainty quantification in machine-assisted medical decision making

Citation Formats

References (21)

Cited By (14)

Similar Records

Related Subjects