Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible.
Established guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal algorithms that confer maximum potential benefits to patients, clinicians, and investigators. We propose a framework for ideal algorithms, in...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLOS Digital Health |
| Online Access: | https://doi.org/10.1371/journal.pdig.0000006 |
| _version_ | 1797700967979810816 |
|---|---|
| author | Tyler J Loftus Patrick J Tighe Tezcan Ozrazgat-Baslanti John P Davis Matthew M Ruppert Yuanfang Ren Benjamin Shickel Rishikesan Kamaleswaran William R Hogan J Randall Moorman Gilbert R Upchurch Parisa Rashidi Azra Bihorac |
| author_facet | Tyler J Loftus Patrick J Tighe Tezcan Ozrazgat-Baslanti John P Davis Matthew M Ruppert Yuanfang Ren Benjamin Shickel Rishikesan Kamaleswaran William R Hogan J Randall Moorman Gilbert R Upchurch Parisa Rashidi Azra Bihorac |
| author_sort | Tyler J Loftus |
| collection | DOAJ |
| description | Established guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal algorithms that confer maximum potential benefits to patients, clinicians, and investigators. We propose a framework for ideal algorithms, including 6 desiderata: explainable (convey the relative importance of features in determining outputs), dynamic (capture temporal changes in physiologic signals and clinical events), precise (use high-resolution, multimodal data and aptly complex architecture), autonomous (learn with minimal supervision and execute without human input), fair (evaluate and mitigate implicit bias and social inequity), and reproducible (validated externally and prospectively and shared with academic communities). We present an ideal algorithms checklist and apply it to highly cited algorithms. Strategies and tools such as the predictive, descriptive, relevant (PDR) framework, the Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence (SPIRIT-AI) extension, sparse regression methods, and minimizing concept drift can help healthcare algorithms achieve these objectives, toward ideal algorithms in healthcare. |
| first_indexed | 2024-03-12T04:28:32Z |
| format | Article |
| id | doaj.art-2b0e333fb5e549789eb87876a49c41eb |
| institution | Directory Open Access Journal |
| issn | 2767-3170 |
| language | English |
| last_indexed | 2024-03-12T04:28:32Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLOS Digital Health |
| spelling | doaj.art-2b0e333fb5e549789eb87876a49c41eb2023-09-03T10:13:41ZengPublic Library of Science (PLoS)PLOS Digital Health2767-31702022-01-0111e000000610.1371/journal.pdig.0000006Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible.Tyler J LoftusPatrick J TigheTezcan Ozrazgat-BaslantiJohn P DavisMatthew M RuppertYuanfang RenBenjamin ShickelRishikesan KamaleswaranWilliam R HoganJ Randall MoormanGilbert R UpchurchParisa RashidiAzra BihoracEstablished guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal algorithms that confer maximum potential benefits to patients, clinicians, and investigators. We propose a framework for ideal algorithms, including 6 desiderata: explainable (convey the relative importance of features in determining outputs), dynamic (capture temporal changes in physiologic signals and clinical events), precise (use high-resolution, multimodal data and aptly complex architecture), autonomous (learn with minimal supervision and execute without human input), fair (evaluate and mitigate implicit bias and social inequity), and reproducible (validated externally and prospectively and shared with academic communities). We present an ideal algorithms checklist and apply it to highly cited algorithms. Strategies and tools such as the predictive, descriptive, relevant (PDR) framework, the Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence (SPIRIT-AI) extension, sparse regression methods, and minimizing concept drift can help healthcare algorithms achieve these objectives, toward ideal algorithms in healthcare.https://doi.org/10.1371/journal.pdig.0000006 |
| spellingShingle | Tyler J Loftus Patrick J Tighe Tezcan Ozrazgat-Baslanti John P Davis Matthew M Ruppert Yuanfang Ren Benjamin Shickel Rishikesan Kamaleswaran William R Hogan J Randall Moorman Gilbert R Upchurch Parisa Rashidi Azra Bihorac Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. PLOS Digital Health |
| title | Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. |
| title_full | Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. |
| title_fullStr | Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. |
| title_full_unstemmed | Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. |
| title_short | Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. |
| title_sort | ideal algorithms in healthcare explainable dynamic precise autonomous fair and reproducible |
| url | https://doi.org/10.1371/journal.pdig.0000006 |
| work_keys_str_mv | AT tylerjloftus idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT patrickjtighe idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT tezcanozrazgatbaslanti idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT johnpdavis idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT matthewmruppert idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT yuanfangren idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT benjaminshickel idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT rishikesankamaleswaran idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT williamrhogan idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT jrandallmoorman idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT gilbertrupchurch idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT parisarashidi idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible AT azrabihorac idealalgorithmsinhealthcareexplainabledynamicpreciseautonomousfairandreproducible |