Raising awareness of potential biases in medical machine learning: Experience from a Datathon.
<h4>Objective</h4>To challenge clinicians and informaticians to learn about potential sources of bias in medical machine learning models through investigation of data and predictions from an open-source severity of illness score.<h4>Methods</h4>Over a two-day period (total el...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2025-07-01
|
| Series: | PLOS Digital Health |
| Online Access: | https://doi.org/10.1371/journal.pdig.0000932 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849316856675958784 |
|---|---|
| author | Harry Hochheiser Jesse Klug Thomas Mathie Tom J Pollard Jesse D Raffa Stephanie L Ballard Evamarie A Conrad Smitha Edakalavan Allan Joseph Nader Alnomasy Sarah Nutman Veronika Hill Sumit Kapoor Eddie Pérez Claudio Olga V Kravchenko Ruoting Li Mehdi Nourelahi Jenny Diaz W Michael Taylor Sydney R Rooney Maeve Woeltje Leo Anthony Celi Christopher M Horvat |
| author_facet | Harry Hochheiser Jesse Klug Thomas Mathie Tom J Pollard Jesse D Raffa Stephanie L Ballard Evamarie A Conrad Smitha Edakalavan Allan Joseph Nader Alnomasy Sarah Nutman Veronika Hill Sumit Kapoor Eddie Pérez Claudio Olga V Kravchenko Ruoting Li Mehdi Nourelahi Jenny Diaz W Michael Taylor Sydney R Rooney Maeve Woeltje Leo Anthony Celi Christopher M Horvat |
| author_sort | Harry Hochheiser |
| collection | DOAJ |
| description | <h4>Objective</h4>To challenge clinicians and informaticians to learn about potential sources of bias in medical machine learning models through investigation of data and predictions from an open-source severity of illness score.<h4>Methods</h4>Over a two-day period (total elapsed time approximately 28 hours), we conducted a datathon that challenged interdisciplinary teams to investigate potential sources of bias in the Global Open Source Severity of Illness Score. Teams were invited to develop hypotheses, to use tools of their choosing to identify potential sources of bias, and to provide a final report.<h4>Results</h4>Five teams participated, three of which included both informaticians and clinicians. Most (4/5) used Python for analyses, the remaining team used R. Common analysis themes included relationship of the GOSSIS-1 prediction score with demographics and care related variables; relationships between demographics and outcomes; calibration and factors related to the context of care; and the impact of missingness. Representativeness of the population, differences in calibration and model performance among groups, and differences in performance across hospital settings were identified as possible sources of bias.<h4>Discussion</h4>Datathons are a promising approach for challenging developers and users to explore questions relating to unrecognized biases in medical machine learning algorithms. |
| format | Article |
| id | doaj-art-a8b7bf778a6e4a2ea124beeafc65ffc2 |
| institution | Kabale University |
| issn | 2767-3170 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLOS Digital Health |
| spelling | doaj-art-a8b7bf778a6e4a2ea124beeafc65ffc22025-08-20T03:51:25ZengPublic Library of Science (PLoS)PLOS Digital Health2767-31702025-07-0147e000093210.1371/journal.pdig.0000932Raising awareness of potential biases in medical machine learning: Experience from a Datathon.Harry HochheiserJesse KlugThomas MathieTom J PollardJesse D RaffaStephanie L BallardEvamarie A ConradSmitha EdakalavanAllan JosephNader AlnomasySarah NutmanVeronika HillSumit KapoorEddie Pérez ClaudioOlga V KravchenkoRuoting LiMehdi NourelahiJenny DiazW Michael TaylorSydney R RooneyMaeve WoeltjeLeo Anthony CeliChristopher M Horvat<h4>Objective</h4>To challenge clinicians and informaticians to learn about potential sources of bias in medical machine learning models through investigation of data and predictions from an open-source severity of illness score.<h4>Methods</h4>Over a two-day period (total elapsed time approximately 28 hours), we conducted a datathon that challenged interdisciplinary teams to investigate potential sources of bias in the Global Open Source Severity of Illness Score. Teams were invited to develop hypotheses, to use tools of their choosing to identify potential sources of bias, and to provide a final report.<h4>Results</h4>Five teams participated, three of which included both informaticians and clinicians. Most (4/5) used Python for analyses, the remaining team used R. Common analysis themes included relationship of the GOSSIS-1 prediction score with demographics and care related variables; relationships between demographics and outcomes; calibration and factors related to the context of care; and the impact of missingness. Representativeness of the population, differences in calibration and model performance among groups, and differences in performance across hospital settings were identified as possible sources of bias.<h4>Discussion</h4>Datathons are a promising approach for challenging developers and users to explore questions relating to unrecognized biases in medical machine learning algorithms.https://doi.org/10.1371/journal.pdig.0000932 |
| spellingShingle | Harry Hochheiser Jesse Klug Thomas Mathie Tom J Pollard Jesse D Raffa Stephanie L Ballard Evamarie A Conrad Smitha Edakalavan Allan Joseph Nader Alnomasy Sarah Nutman Veronika Hill Sumit Kapoor Eddie Pérez Claudio Olga V Kravchenko Ruoting Li Mehdi Nourelahi Jenny Diaz W Michael Taylor Sydney R Rooney Maeve Woeltje Leo Anthony Celi Christopher M Horvat Raising awareness of potential biases in medical machine learning: Experience from a Datathon. PLOS Digital Health |
| title | Raising awareness of potential biases in medical machine learning: Experience from a Datathon. |
| title_full | Raising awareness of potential biases in medical machine learning: Experience from a Datathon. |
| title_fullStr | Raising awareness of potential biases in medical machine learning: Experience from a Datathon. |
| title_full_unstemmed | Raising awareness of potential biases in medical machine learning: Experience from a Datathon. |
| title_short | Raising awareness of potential biases in medical machine learning: Experience from a Datathon. |
| title_sort | raising awareness of potential biases in medical machine learning experience from a datathon |
| url | https://doi.org/10.1371/journal.pdig.0000932 |
| work_keys_str_mv | AT harryhochheiser raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT jesseklug raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT thomasmathie raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT tomjpollard raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT jessedraffa raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT stephanielballard raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT evamarieaconrad raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT smithaedakalavan raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT allanjoseph raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT naderalnomasy raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT sarahnutman raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT veronikahill raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT sumitkapoor raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT eddieperezclaudio raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT olgavkravchenko raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT ruotingli raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT mehdinourelahi raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT jennydiaz raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT wmichaeltaylor raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT sydneyrrooney raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT maevewoeltje raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT leoanthonyceli raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon AT christophermhorvat raisingawarenessofpotentialbiasesinmedicalmachinelearningexperiencefromadatathon |