Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes
Abstract Background Being able to predict with confidence the early onset of type 2 diabetes from a suite of signs and symptoms (features) displayed by potential sufferers is desirable to commence treatment promptly. Late or inconclusive diagnosis can result in more serious health consequences for s...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-12-01
|
| Series: | Chronic Diseases and Translational Medicine |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/cdt3.39 |
| _version_ | 1797733422171422720 |
|---|---|
| author | David A. Wood |
| author_facet | David A. Wood |
| author_sort | David A. Wood |
| collection | DOAJ |
| description | Abstract Background Being able to predict with confidence the early onset of type 2 diabetes from a suite of signs and symptoms (features) displayed by potential sufferers is desirable to commence treatment promptly. Late or inconclusive diagnosis can result in more serious health consequences for sufferers and higher costs for health care services in the long run. Methods A novel integrated methodology is proposed involving correlation, statistical analysis, machine learning, multi‐K‐fold cross‐validation, and confusion matrices to provide a reliable classification of diabetes‐positive and ‐negative individuals from a substantial suite of features. The method also identifies the relative influence of each feature on the diabetes diagnosis and highlights the most important ones. Ten statistical and machine learning methods are utilized to conduct the analysis. Results A published data set involving 520 individuals (Sylthet Diabetes Hospital, Bangladesh) is modeled revealing that a support vector classifier generates the most accurate early‐onset type 2 diabetes status predictions with just 11 misclassifications (2.1% error). Polydipsia and polyuria are among the most influential features, whereas obesity and age are assigned low weights by the prediction models. Conclusion The proposed methodology can rapidly predict early‐onset type 2 diabetes with high confidence while providing valuable insight into the key influential features involved in such predictions. |
| first_indexed | 2024-03-12T12:28:51Z |
| format | Article |
| id | doaj.art-9294d69bf9c14640acb37c83dc8aa40d |
| institution | Directory Open Access Journal |
| issn | 2589-0514 |
| language | English |
| last_indexed | 2024-03-12T12:28:51Z |
| publishDate | 2022-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | Chronic Diseases and Translational Medicine |
| spelling | doaj.art-9294d69bf9c14640acb37c83dc8aa40d2023-08-29T18:05:52ZengWileyChronic Diseases and Translational Medicine2589-05142022-12-018428129510.1002/cdt3.39Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetesDavid A. Wood0DWA Energy Limited Lincoln UKAbstract Background Being able to predict with confidence the early onset of type 2 diabetes from a suite of signs and symptoms (features) displayed by potential sufferers is desirable to commence treatment promptly. Late or inconclusive diagnosis can result in more serious health consequences for sufferers and higher costs for health care services in the long run. Methods A novel integrated methodology is proposed involving correlation, statistical analysis, machine learning, multi‐K‐fold cross‐validation, and confusion matrices to provide a reliable classification of diabetes‐positive and ‐negative individuals from a substantial suite of features. The method also identifies the relative influence of each feature on the diabetes diagnosis and highlights the most important ones. Ten statistical and machine learning methods are utilized to conduct the analysis. Results A published data set involving 520 individuals (Sylthet Diabetes Hospital, Bangladesh) is modeled revealing that a support vector classifier generates the most accurate early‐onset type 2 diabetes status predictions with just 11 misclassifications (2.1% error). Polydipsia and polyuria are among the most influential features, whereas obesity and age are assigned low weights by the prediction models. Conclusion The proposed methodology can rapidly predict early‐onset type 2 diabetes with high confidence while providing valuable insight into the key influential features involved in such predictions.https://doi.org/10.1002/cdt3.39error analysiskey feature influencesmulti‐K‐fold cross‐validationsymptom importancetype 2 diabetes screening |
| spellingShingle | David A. Wood Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes Chronic Diseases and Translational Medicine error analysis key feature influences multi‐K‐fold cross‐validation symptom importance type 2 diabetes screening |
| title | Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes |
| title_full | Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes |
| title_fullStr | Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes |
| title_full_unstemmed | Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes |
| title_short | Integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early‐onset type 2 diabetes |
| title_sort | integrated statistical and machine learning analysis provides insight into key influencing symptoms for distinguishing early onset type 2 diabetes |
| topic | error analysis key feature influences multi‐K‐fold cross‐validation symptom importance type 2 diabetes screening |
| url | https://doi.org/10.1002/cdt3.39 |
| work_keys_str_mv | AT davidawood integratedstatisticalandmachinelearninganalysisprovidesinsightintokeyinfluencingsymptomsfordistinguishingearlyonsettype2diabetes |