Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining
Reliably detecting potentially misleading patterns in automated diagnostic assistance systems, such as those powered by artificial intelligence (AI), is crucial for instilling user trust and ensuring reliability. Current techniques fall short in visualizing such confounding factors. We propose DiffC...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
17 September 2024
|
| In: |
Cell reports. Medicine
Year: 2024, Volume: 5, Issue: 9, Pages: 1-11 |
| ISSN: | 2666-3791 |
| DOI: | 10.1016/j.xcrm.2024.101713 |
| Online Access: | Verlag, kostenfrei, Volltext: https://doi.org/10.1016/j.xcrm.2024.101713 Verlag, kostenfrei, Volltext: https://www.sciencedirect.com/science/article/pii/S2666379124004348 
|
| Author Notes: | Tianyu Han, Laura Žigutytė, Luisa Huck, Marc Sebastian Huppertz, Robert Siepmann, Yossi Gandelsman, Christian Blüthgen, Firas Khader, Christiane Kuhl, Sven Nebelung, Jakob Nikolas Kather, Daniel Truhn |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1923049844 | ||
| 003 | DE-627 | ||
| 005 | 20250717010416.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 250416s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.xcrm.2024.101713 |2 doi | |
| 035 | |a (DE-627)1923049844 | ||
| 035 | |a (DE-599)KXP1923049844 | ||
| 035 | |a (OCoLC)1528044681 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Han, Tianyu |e VerfasserIn |0 (DE-588)1278510001 |0 (DE-627)1831430037 |4 aut | |
| 245 | 1 | 0 | |a Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining |c Tianyu Han, Laura Žigutytė, Luisa Huck, Marc Sebastian Huppertz, Robert Siepmann, Yossi Gandelsman, Christian Blüthgen, Firas Khader, Christiane Kuhl, Sven Nebelung, Jakob Nikolas Kather, Daniel Truhn |
| 264 | 1 | |c 17 September 2024 | |
| 300 | |b Illustrationen, Diagramme | ||
| 300 | |a 18 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 16.04.2025 | ||
| 520 | |a Reliably detecting potentially misleading patterns in automated diagnostic assistance systems, such as those powered by artificial intelligence (AI), is crucial for instilling user trust and ensuring reliability. Current techniques fall short in visualizing such confounding factors. We propose DiffChest, a self-conditioned diffusion model trained on 515,704 chest radiographs from 194,956 patients across the US and Europe. DiffChest provides patient-specific explanations and visualizes confounding factors that might mislead the model. The high inter-reader agreement, with Fleiss’ kappa values of 0.8 or higher, validates its capability to identify treatment-related confounders. Confounders are accurately detected with 10%-100% prevalence rates. The pretraining process optimizes the model for relevant imaging information, resulting in excellent diagnostic accuracy for 11 chest conditions, including pleural effusion and heart insufficiency. Our findings highlight the potential of diffusion models in medical image classification, providing insights into confounding factors and enhancing model robustness and reliability. | ||
| 650 | 4 | |a confounders | |
| 650 | 4 | |a counterfactual explanations | |
| 650 | 4 | |a deep learning | |
| 650 | 4 | |a explainability | |
| 650 | 4 | |a generative models | |
| 650 | 4 | |a medical imaging | |
| 650 | 4 | |a self-supervised training | |
| 700 | 1 | |a Žigutytė, Laura |e VerfasserIn |4 aut | |
| 700 | 1 | |a Huck, Luisa |e VerfasserIn |4 aut | |
| 700 | 1 | |a Huppertz, Marc Sebastian |e VerfasserIn |4 aut | |
| 700 | 1 | |a Siepmann, Robert |e VerfasserIn |4 aut | |
| 700 | 1 | |a Gandelsman, Yossi |e VerfasserIn |4 aut | |
| 700 | 1 | |a Blüthgen, Christian |e VerfasserIn |4 aut | |
| 700 | 1 | |a Khader, Firas |e VerfasserIn |4 aut | |
| 700 | 1 | |8 1\p |a Kuhl, Christiane |d 1966- |e VerfasserIn |0 (DE-588)1082384011 |0 (DE-627)847427994 |0 (DE-576)455377782 |4 aut | |
| 700 | 1 | |a Nebelung, Sven |e VerfasserIn |4 aut | |
| 700 | 1 | |a Kather, Jakob Nikolas |d 1989- |e VerfasserIn |0 (DE-588)1064064914 |0 (DE-627)812897587 |0 (DE-576)423589091 |4 aut | |
| 700 | 1 | |8 2\p |a Truhn, Daniel |e VerfasserIn |0 (DE-588)1047348306 |0 (DE-627)778145913 |0 (DE-576)400927314 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Cell reports. Medicine |d Cambridge, MA : Cell Press, 2020 |g Volume 5, issue 9 (17 September 2024), article no. 101713, 1 ungezählte Seite, 11 Seiten, Seite e1-e6 |h Online-Ressource |w (DE-627)1696877792 |w (DE-600)3019420-9 |x 2666-3791 |7 nnas |a Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining |
| 773 | 1 | 8 | |g volume:5 |g year:2024 |g number:9 |g month:09 |g elocationid:101713 |g pages:1-11 |g extent:18 |a Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining |
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.xcrm.2024.101713 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u https://www.sciencedirect.com/science/article/pii/S2666379124004348 |x Verlag |z kostenfrei |3 Volltext |
| 883 | |8 1\p |a cgwrk |d 20250505 |q DE-101 |u https://d-nb.info/provenance/plan#cgwrk | ||
| 883 | |8 2\p |a cgwrk |d 20250505 |q DE-101 |u https://d-nb.info/provenance/plan#cgwrk | ||
| 951 | |a AR | ||
| 992 | |a 20250416 | ||
| 993 | |a Article | ||
| 994 | |a 2024 | ||
| 998 | |g 1064064914 |a Kather, Jakob Nikolas |m 1064064914:Kather, Jakob Nikolas |d 910000 |d 910100 |e 910000PK1064064914 |e 910100PK1064064914 |k 0/910000/ |k 1/910000/910100/ |p 11 | ||
| 999 | |a KXP-PPN1923049844 |e 4706002877 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"name":{"displayForm":["Tianyu Han, Laura Žigutytė, Luisa Huck, Marc Sebastian Huppertz, Robert Siepmann, Yossi Gandelsman, Christian Blüthgen, Firas Khader, Christiane Kuhl, Sven Nebelung, Jakob Nikolas Kather, Daniel Truhn"]},"recId":"1923049844","language":["eng"],"note":["Gesehen am 16.04.2025"],"origin":[{"dateIssuedKey":"2024","dateIssuedDisp":"17 September 2024"}],"id":{"doi":["10.1016/j.xcrm.2024.101713"],"eki":["1923049844"]},"title":[{"title_sort":"Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining","title":"Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining"}],"relHost":[{"disp":"Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretrainingCell reports. Medicine","origin":[{"publisherPlace":"Cambridge, MA ; Maryland Heights, MO","publisher":"Cell Press ; Elsevier","dateIssuedDisp":"[2020]-"}],"language":["eng"],"part":{"issue":"9","pages":"1-11","text":"Volume 5, issue 9 (17 September 2024), article no. 101713, 1 ungezählte Seite, 11 Seiten, Seite e1-e6","year":"2024","volume":"5","extent":"18"},"note":["Gesehen am 29. April 2020"],"recId":"1696877792","type":{"bibl":"periodical","media":"Online-Ressource"},"physDesc":[{"extent":"Online-Ressource"}],"title":[{"partname":"Medicine","title":"Cell reports","title_sort":"Cell reports"}],"id":{"eki":["1696877792"],"issn":["2666-3791"],"zdb":["3019420-9"]},"pubHistory":["Volume 1, issue 1 (2020)-"]}],"physDesc":[{"extent":"18 S.","noteIll":"Illustrationen, Diagramme"}],"type":{"bibl":"article-journal","media":"Online-Ressource"},"person":[{"display":"Han, Tianyu","family":"Han","given":"Tianyu","role":"aut"},{"role":"aut","given":"Laura","family":"Žigutytė","display":"Žigutytė, Laura"},{"family":"Huck","display":"Huck, Luisa","role":"aut","given":"Luisa"},{"given":"Marc Sebastian","role":"aut","display":"Huppertz, Marc Sebastian","family":"Huppertz"},{"role":"aut","given":"Robert","family":"Siepmann","display":"Siepmann, Robert"},{"role":"aut","given":"Yossi","family":"Gandelsman","display":"Gandelsman, Yossi"},{"display":"Blüthgen, Christian","family":"Blüthgen","given":"Christian","role":"aut"},{"display":"Khader, Firas","family":"Khader","given":"Firas","role":"aut"},{"family":"Kuhl","display":"Kuhl, Christiane","role":"aut","given":"Christiane"},{"given":"Sven","role":"aut","display":"Nebelung, Sven","family":"Nebelung"},{"display":"Kather, Jakob Nikolas","family":"Kather","given":"Jakob Nikolas","role":"aut"},{"display":"Truhn, Daniel","family":"Truhn","given":"Daniel","role":"aut"}]} | ||
| SRT | |a HANTIANYUZRECONSTRUC1720 | ||