WorkDifferentWithAI.com Academic Article Alert!
Written by Judy Wawira Gichoya, Kaesha Thomas, Leo Anthony Celi, Nabile Safdar, Imon Banerjee, John D Banja, Laleh Seyyed-Kalantari, Hari Trivedi, Saptarshi Purkayastha
Category: “AI Strategy & Governance”
Article Section: Ethics of enterprise AI
Publication Date: 2023-10
SEO Description: “Managing Bias in AI for Healthcare: Issues & Mitigation Strategies”
12014432
{12014432:CD52FAWN}
items
1
modern-language-association
0
default
asc
530
https://workdifferentwithai.net/wp-content/plugins/zotpress/
%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3A%22zotpress-dab924c24548561161a735dafc4dfa0f%22%2C%22meta%22%3A%7B%22request_last%22%3A0%2C%22request_next%22%3A0%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%22CD52FAWN%22%2C%22library%22%3A%7B%22id%22%3A12014432%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Gichoya%20et%20al.%22%2C%22parsedDate%22%3A%222023-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%202%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%3EGichoya%2C%20Judy%20Wawira%2C%20et%20al.%20%26%23x201C%3BAI%20Pitfalls%20and%20What%20Not%20to%20Do%3A%20Mitigating%20Bias%20in%20AI.%26%23x201D%3B%20%3Ci%3EThe%20British%20Journal%20of%20Radiology%3C%5C%2Fi%3E%2C%20vol.%2096%2C%20no.%201150%2C%20Oct.%202023%2C%20p.%2020230023%2C%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1259%5C%2Fbjr.20230023%27%3Ehttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1259%5C%2Fbjr.20230023%3C%5C%2Fa%3E.%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22AI%20pitfalls%20and%20what%20not%20to%20do%3A%20mitigating%20bias%20in%20AI%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Judy%20Wawira%22%2C%22lastName%22%3A%22Gichoya%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kaesha%22%2C%22lastName%22%3A%22Thomas%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Leo%20Anthony%22%2C%22lastName%22%3A%22Celi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Nabile%22%2C%22lastName%22%3A%22Safdar%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Imon%22%2C%22lastName%22%3A%22Banerjee%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22John%20D%22%2C%22lastName%22%3A%22Banja%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Laleh%22%2C%22lastName%22%3A%22Seyyed-Kalantari%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hari%22%2C%22lastName%22%3A%22Trivedi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Saptarshi%22%2C%22lastName%22%3A%22Purkayastha%22%7D%5D%2C%22abstractNote%22%3A%22Various%20forms%20of%20artificial%20intelligence%20%28AI%29%20applications%20are%20being%20deployed%20and%20used%20in%20many%20healthcare%20systems.%20As%20the%20use%20of%20these%20applications%20increases%2C%20we%20are%20learning%20the%20failures%20of%20these%20models%20and%20how%20they%20can%20perpetuate%20bias.%20With%20these%20new%20lessons%2C%20we%20need%20to%20prioritize%20bias%20evaluation%20and%20mitigation%20for%20radiology%20applications%3B%20all%20the%20while%20not%20ignoring%20the%20impact%20of%20changes%20in%20the%20larger%20enterprise%20AI%20deployment%20which%20may%20have%20downstream%20impact%20on%20performance%20of%20AI%20models.%20In%20this%20paper%2C%20we%20provide%20an%20updated%20review%20of%20known%20pitfalls%20causing%20AI%20bias%20and%20discuss%20strategies%20for%20mitigating%20these%20biases%20within%20the%20context%20of%20AI%20deployment%20in%20the%20larger%20healthcare%20enterprise.%20We%20describe%20these%20pitfalls%20by%20framing%20them%20in%20the%20larger%20AI%20lifecycle%20from%20problem%20definition%2C%20data%20set%20selection%20and%20curation%2C%20model%20training%20and%20deployment%20emphasizing%20that%20bias%20exists%20across%20a%20spectrum%20and%20is%20a%20sequela%20of%20a%20combination%20of%20both%20human%20and%20machine%20factors.%22%2C%22date%22%3A%222023-10%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1259%5C%2Fbjr.20230023%22%2C%22ISSN%22%3A%220007-1285%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.birpublications.org%5C%2Fdoi%5C%2Ffull%5C%2F10.1259%5C%2Fbjr.20230023%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222023-10-18T01%3A35%3A11Z%22%7D%7D%5D%7D
Gichoya, Judy Wawira, et al. “AI Pitfalls and What Not to Do: Mitigating Bias in AI.”
The British Journal of Radiology, vol. 96, no. 1150, Oct. 2023, p. 20230023,
https://doi.org/10.1259/bjr.20230023.
GPT Generated Article Summary
This text discusses the application of artificial intelligence (AI) in healthcare systems and the potential pitfalls relating to bias that can arise from these models. As the usage of AI increases in healthcare, biases in these models can be unintentionally perpetuated. Hence, it underscores the urgency to prioritize evaluating and mitigating this bias in radiology applications, without overlooking possible effects from broader enterprise AI deployment. These challenges are discussed within the larger context of the AI lifecycle which spans problem definition, dataset selection, curation, and model training. It emphasises that bias exists across a spectrum and is a result of both human and machine factors.
Keywords
AI pitfalls, bias mitigation, healthcare systems, radiology applications, AI deployment
Author’s Abstract
Various forms of artificial intelligence (AI) applications are being deployed and used in many healthcare systems. As the use of these applications increases, we are learning the failures of these models and how they can perpetuate bias. With these new lessons, we need to prioritize bias evaluation and mitigation for radiology applications; all the while not ignoring the impact of changes in the larger enterprise AI deployment which may have downstream impact on performance of AI models. In this paper, we provide an updated review of known pitfalls causing AI bias and discuss strategies for mitigating these biases within the context of AI deployment in the larger healthcare enterprise. We describe these pitfalls by framing them in the larger AI lifecycle from problem definition, data set selection and curation, model training and deployment emphasizing that bias exists across a spectrum and is a sequela of a combination of both human and machine factors.
Read the full article here
