Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study
Screening tests for disease have their performance measured through sensitivity and specificity, which inform how well the test can discriminate between those with and without the condition. Typically, high values for sensitivity and specificity are desired. These two measures of performance are una...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Experimental Biology and Medicine |
| Subjects: | |
| Online Access: | https://www.ebm-journal.org/articles/10.3389/ebm.2024.10341/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841555731649658880 |
|---|---|
| author | Paul Rogers Thomas McCall Ying Zhang Jessica Reese Dong Wang Weida Tong |
| author_facet | Paul Rogers Thomas McCall Ying Zhang Jessica Reese Dong Wang Weida Tong |
| author_sort | Paul Rogers |
| collection | DOAJ |
| description | Screening tests for disease have their performance measured through sensitivity and specificity, which inform how well the test can discriminate between those with and without the condition. Typically, high values for sensitivity and specificity are desired. These two measures of performance are unaffected by the outcome prevalence of the disease in the population. Research projects into the health of the American Indian frequently develop Machine learning algorithms as predictors of conditions in this population. In essence, these models serve as in silico screening tests for disease. A screening test’s sensitivity and specificity values, typically determined during the development of the test, inform on the performance at the population level and are not affected by the prevalence of disease. A screening test’s positive predictive value (PPV) is susceptible to the prevalence of the outcome. As the number of artificial intelligence and machine learning models flourish to predict disease outcomes, it is crucial to understand if the PPV values for these in silico methods suffer as traditional screening tests in a low prevalence outcome environment. The Strong Heart Study (SHS) is an epidemiological study of the American Indian and has been utilized in predictive models for health outcomes. We used data from the SHS focusing on the samples taken during Phases V and VI. Logistic Regression, Artificial Neural Network, and Random Forest were utilized as in silico screening tests within the SHS group. Their sensitivity, specificity, and PPV performance were assessed with health outcomes of varying prevalence within the SHS subjects. Although sensitivity and specificity remained high in these in silico screening tests, the PPVs’ values declined as the outcome’s prevalence became rare. Machine learning models used as in silico screening tests are subject to the same drawbacks as traditional screening tests when the outcome to be predicted is of low prevalence. |
| format | Article |
| id | doaj-art-28768d24464c42449f1ceeb15301115e |
| institution | Kabale University |
| issn | 1535-3699 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Experimental Biology and Medicine |
| spelling | doaj-art-28768d24464c42449f1ceeb15301115e2025-01-08T04:11:13ZengFrontiers Media S.A.Experimental Biology and Medicine1535-36992025-01-0124910.3389/ebm.2024.1034110341Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart StudyPaul Rogers0Thomas McCall1Ying Zhang2Jessica Reese3Dong Wang4Weida Tong5National Center for Toxicological Research, Division of Bioinformatics and Biostatistics, U.S. Food and Drug Administration, Jefferson, AR, United StatesDepartment of Data Science and Data Analytics, Arkansas State University, Jonesboro, AR, United StatesUniversity of Oklahoma Health Sciences Center, Department of Biostatistics and Epidemiology, Oklahoma City, OK, United StatesUniversity of Oklahoma Health Sciences Center, Department of Biostatistics and Epidemiology, Oklahoma City, OK, United StatesNational Center for Toxicological Research, Division of Bioinformatics and Biostatistics, U.S. Food and Drug Administration, Jefferson, AR, United StatesNational Center for Toxicological Research, Division of Bioinformatics and Biostatistics, U.S. Food and Drug Administration, Jefferson, AR, United StatesScreening tests for disease have their performance measured through sensitivity and specificity, which inform how well the test can discriminate between those with and without the condition. Typically, high values for sensitivity and specificity are desired. These two measures of performance are unaffected by the outcome prevalence of the disease in the population. Research projects into the health of the American Indian frequently develop Machine learning algorithms as predictors of conditions in this population. In essence, these models serve as in silico screening tests for disease. A screening test’s sensitivity and specificity values, typically determined during the development of the test, inform on the performance at the population level and are not affected by the prevalence of disease. A screening test’s positive predictive value (PPV) is susceptible to the prevalence of the outcome. As the number of artificial intelligence and machine learning models flourish to predict disease outcomes, it is crucial to understand if the PPV values for these in silico methods suffer as traditional screening tests in a low prevalence outcome environment. The Strong Heart Study (SHS) is an epidemiological study of the American Indian and has been utilized in predictive models for health outcomes. We used data from the SHS focusing on the samples taken during Phases V and VI. Logistic Regression, Artificial Neural Network, and Random Forest were utilized as in silico screening tests within the SHS group. Their sensitivity, specificity, and PPV performance were assessed with health outcomes of varying prevalence within the SHS subjects. Although sensitivity and specificity remained high in these in silico screening tests, the PPVs’ values declined as the outcome’s prevalence became rare. Machine learning models used as in silico screening tests are subject to the same drawbacks as traditional screening tests when the outcome to be predicted is of low prevalence.https://www.ebm-journal.org/articles/10.3389/ebm.2024.10341/fullartificial intelligencemachine learningscreening testAmerican Indianlow prevalence |
| spellingShingle | Paul Rogers Thomas McCall Ying Zhang Jessica Reese Dong Wang Weida Tong Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study Experimental Biology and Medicine artificial intelligence machine learning screening test American Indian low prevalence |
| title | Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study |
| title_full | Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study |
| title_fullStr | Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study |
| title_full_unstemmed | Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study |
| title_short | Leveraging AI to improve disease screening among American Indians: insights from the Strong Heart Study |
| title_sort | leveraging ai to improve disease screening among american indians insights from the strong heart study |
| topic | artificial intelligence machine learning screening test American Indian low prevalence |
| url | https://www.ebm-journal.org/articles/10.3389/ebm.2024.10341/full |
| work_keys_str_mv | AT paulrogers leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy AT thomasmccall leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy AT yingzhang leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy AT jessicareese leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy AT dongwang leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy AT weidatong leveragingaitoimprovediseasescreeningamongamericanindiansinsightsfromthestrongheartstudy |