Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis
Abstract Aims/Introduction Recently, an increasing number of cohort studies have suggested using machine learning (ML) to predict type 2 diabetes mellitus. However, its predictive ability remains inconclusive. This meta‐analysis evaluated the current ability of ML algorithms for predicting incident...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-05-01
|
| Series: | Journal of Diabetes Investigation |
| Subjects: | |
| Online Access: | https://doi.org/10.1111/jdi.13736 |
| _version_ | 1828351045789548544 |
|---|---|
| author | Satoru Kodama Kazuya Fujihara Chika Horikawa Masaru Kitazawa Midori Iwanaga Kiminori Kato Kenichi Watanabe Yoshimi Nakagawa Takashi Matsuzaka Hitoshi Shimano Hirohito Sone |
| author_facet | Satoru Kodama Kazuya Fujihara Chika Horikawa Masaru Kitazawa Midori Iwanaga Kiminori Kato Kenichi Watanabe Yoshimi Nakagawa Takashi Matsuzaka Hitoshi Shimano Hirohito Sone |
| author_sort | Satoru Kodama |
| collection | DOAJ |
| description | Abstract Aims/Introduction Recently, an increasing number of cohort studies have suggested using machine learning (ML) to predict type 2 diabetes mellitus. However, its predictive ability remains inconclusive. This meta‐analysis evaluated the current ability of ML algorithms for predicting incident type 2 diabetes mellitus. Materials and Methods We systematically searched longitudinal studies published from 1 January 1950 to 17 May 2020 using MEDLINE and EMBASE. Included studies had to compare ML’s classification with the actual incidence of type 2 diabetes mellitus, and present data on the number of true positives, false positives, true negatives and false negatives. The dataset for these four values was pooled with a hierarchical summary receiver operating characteristic and a bivariate random effects model. Results There were 12 eligible studies. The pooled sensitivity, specificity, positive likelihood ratio and negative likelihood ratio were 0.81 (95% confidence interval [CI] 0.67–0.90), 0.82 [95% CI 0.74–0.88], 4.55 [95% CI 3.07–6.75] and 0.23 [95% CI 0.13–0.42], respectively. The area under the summarized receiver operating characteristic curve was 0.88 (95% CI 0.85–0.91). Conclusions Current ML algorithms have sufficient ability to help clinicians determine whether individuals will develop type 2 diabetes mellitus in the future. However, persons should be cautious before changing their attitude toward future diabetes risk after learning the result of the diabetes prediction test using ML algorithms. |
| first_indexed | 2024-04-14T01:33:53Z |
| format | Article |
| id | doaj.art-5b000a41fb1148178bb79a1c1e8ebd16 |
| institution | Directory Open Access Journal |
| issn | 2040-1116 2040-1124 |
| language | English |
| last_indexed | 2024-04-14T01:33:53Z |
| publishDate | 2022-05-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of Diabetes Investigation |
| spelling | doaj.art-5b000a41fb1148178bb79a1c1e8ebd162022-12-22T02:20:03ZengWileyJournal of Diabetes Investigation2040-11162040-11242022-05-0113590090810.1111/jdi.13736Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysisSatoru Kodama0Kazuya Fujihara1Chika Horikawa2Masaru Kitazawa3Midori Iwanaga4Kiminori Kato5Kenichi Watanabe6Yoshimi Nakagawa7Takashi Matsuzaka8Hitoshi Shimano9Hirohito Sone10Department of Prevention of Noncommunicable Diseases and Promotion of Health Checkup Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDepartment of Hematology, Endocrinology and Metabolism Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDepartment of Health and Nutrition Faculty of Human Life Studies University of Niigata Prefecture Niigata JapanDepartment of Prevention of Noncommunicable Diseases and Promotion of Health Checkup Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDepartment of Prevention of Noncommunicable Diseases and Promotion of Health Checkup Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDepartment of Prevention of Noncommunicable Diseases and Promotion of Health Checkup Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDepartment of Prevention of Noncommunicable Diseases and Promotion of Health Checkup Niigata University Graduate School of Medical and Dental Sciences Niigata JapanDivision of Complex Biosystem Research Institute of Natural Medicine Toyama University Toyama JapanDepartment of Internal Medicine (Endocrinology and Metabolism) Faculty of Medicine University of Tsukuba Ibaraki JapanDepartment of Internal Medicine (Endocrinology and Metabolism) Faculty of Medicine University of Tsukuba Ibaraki JapanDepartment of Hematology, Endocrinology and Metabolism Niigata University Graduate School of Medical and Dental Sciences Niigata JapanAbstract Aims/Introduction Recently, an increasing number of cohort studies have suggested using machine learning (ML) to predict type 2 diabetes mellitus. However, its predictive ability remains inconclusive. This meta‐analysis evaluated the current ability of ML algorithms for predicting incident type 2 diabetes mellitus. Materials and Methods We systematically searched longitudinal studies published from 1 January 1950 to 17 May 2020 using MEDLINE and EMBASE. Included studies had to compare ML’s classification with the actual incidence of type 2 diabetes mellitus, and present data on the number of true positives, false positives, true negatives and false negatives. The dataset for these four values was pooled with a hierarchical summary receiver operating characteristic and a bivariate random effects model. Results There were 12 eligible studies. The pooled sensitivity, specificity, positive likelihood ratio and negative likelihood ratio were 0.81 (95% confidence interval [CI] 0.67–0.90), 0.82 [95% CI 0.74–0.88], 4.55 [95% CI 3.07–6.75] and 0.23 [95% CI 0.13–0.42], respectively. The area under the summarized receiver operating characteristic curve was 0.88 (95% CI 0.85–0.91). Conclusions Current ML algorithms have sufficient ability to help clinicians determine whether individuals will develop type 2 diabetes mellitus in the future. However, persons should be cautious before changing their attitude toward future diabetes risk after learning the result of the diabetes prediction test using ML algorithms.https://doi.org/10.1111/jdi.13736Machine learningMeta‐analysisType 2 diabetes mellitus |
| spellingShingle | Satoru Kodama Kazuya Fujihara Chika Horikawa Masaru Kitazawa Midori Iwanaga Kiminori Kato Kenichi Watanabe Yoshimi Nakagawa Takashi Matsuzaka Hitoshi Shimano Hirohito Sone Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis Journal of Diabetes Investigation Machine learning Meta‐analysis Type 2 diabetes mellitus |
| title | Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis |
| title_full | Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis |
| title_fullStr | Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis |
| title_full_unstemmed | Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis |
| title_short | Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta‐analysis |
| title_sort | predictive ability of current machine learning algorithms for type 2 diabetes mellitus a meta analysis |
| topic | Machine learning Meta‐analysis Type 2 diabetes mellitus |
| url | https://doi.org/10.1111/jdi.13736 |
| work_keys_str_mv | AT satorukodama predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT kazuyafujihara predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT chikahorikawa predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT masarukitazawa predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT midoriiwanaga predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT kiminorikato predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT kenichiwatanabe predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT yoshiminakagawa predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT takashimatsuzaka predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT hitoshishimano predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis AT hirohitosone predictiveabilityofcurrentmachinelearningalgorithmsfortype2diabetesmellitusametaanalysis |