A moment kernel machine for clinical data mining to inform medical decision making
Abstract Machine learning-aided medical decision making presents three major challenges: achieving model parsimony, ensuring credible predictions, and providing real-time recommendations with high computational efficiency. In this paper, we formulate medical decision making as a classification probl...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2023-06-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-023-36752-7 |
_version_ | 1797789839739846656 |
---|---|
author | Yao-Chi Yu Wei Zhang David O’Gara Jr-Shin Li Su-Hsin Chang |
author_facet | Yao-Chi Yu Wei Zhang David O’Gara Jr-Shin Li Su-Hsin Chang |
author_sort | Yao-Chi Yu |
collection | DOAJ |
description | Abstract Machine learning-aided medical decision making presents three major challenges: achieving model parsimony, ensuring credible predictions, and providing real-time recommendations with high computational efficiency. In this paper, we formulate medical decision making as a classification problem and develop a moment kernel machine (MKM) to tackle these challenges. The main idea of our approach is to treat the clinical data of each patient as a probability distribution and leverage moment representations of these distributions to build the MKM, which transforms the high-dimensional clinical data to low-dimensional representations while retaining essential information. We then apply this machine to various pre-surgical clinical datasets to predict surgical outcomes and inform medical decision making, which requires significantly less computational power and time for classification while yielding favorable performance compared to existing methods. Moreover, we utilize synthetic datasets to demonstrate that the developed moment-based data mining framework is robust to noise and missing data, and achieves model parsimony giving an efficient way to generate satisfactory predictions to aid personalized medical decision making. |
first_indexed | 2024-03-13T01:56:17Z |
format | Article |
id | doaj.art-51ebd06be49f4388add167ff6b450474 |
institution | Directory Open Access Journal |
issn | 2045-2322 |
language | English |
last_indexed | 2024-03-13T01:56:17Z |
publishDate | 2023-06-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Scientific Reports |
spelling | doaj.art-51ebd06be49f4388add167ff6b4504742023-07-02T11:12:45ZengNature PortfolioScientific Reports2045-23222023-06-0113111110.1038/s41598-023-36752-7A moment kernel machine for clinical data mining to inform medical decision makingYao-Chi Yu0Wei Zhang1David O’Gara2Jr-Shin Li3Su-Hsin Chang4Department of Electrical and Systems Engineering, Washington University in St. LouisDepartment of Electrical and Systems Engineering, Washington University in St. LouisDivision of Computational and Data Sciences, Washington University in St. LouisDepartment of Electrical and Systems Engineering, Washington University in St. LouisDivision of Public Health Sciences, Department of Surgery, Washington University School of MedicineAbstract Machine learning-aided medical decision making presents three major challenges: achieving model parsimony, ensuring credible predictions, and providing real-time recommendations with high computational efficiency. In this paper, we formulate medical decision making as a classification problem and develop a moment kernel machine (MKM) to tackle these challenges. The main idea of our approach is to treat the clinical data of each patient as a probability distribution and leverage moment representations of these distributions to build the MKM, which transforms the high-dimensional clinical data to low-dimensional representations while retaining essential information. We then apply this machine to various pre-surgical clinical datasets to predict surgical outcomes and inform medical decision making, which requires significantly less computational power and time for classification while yielding favorable performance compared to existing methods. Moreover, we utilize synthetic datasets to demonstrate that the developed moment-based data mining framework is robust to noise and missing data, and achieves model parsimony giving an efficient way to generate satisfactory predictions to aid personalized medical decision making.https://doi.org/10.1038/s41598-023-36752-7 |
spellingShingle | Yao-Chi Yu Wei Zhang David O’Gara Jr-Shin Li Su-Hsin Chang A moment kernel machine for clinical data mining to inform medical decision making Scientific Reports |
title | A moment kernel machine for clinical data mining to inform medical decision making |
title_full | A moment kernel machine for clinical data mining to inform medical decision making |
title_fullStr | A moment kernel machine for clinical data mining to inform medical decision making |
title_full_unstemmed | A moment kernel machine for clinical data mining to inform medical decision making |
title_short | A moment kernel machine for clinical data mining to inform medical decision making |
title_sort | moment kernel machine for clinical data mining to inform medical decision making |
url | https://doi.org/10.1038/s41598-023-36752-7 |
work_keys_str_mv | AT yaochiyu amomentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT weizhang amomentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT davidogara amomentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT jrshinli amomentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT suhsinchang amomentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT yaochiyu momentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT weizhang momentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT davidogara momentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT jrshinli momentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking AT suhsinchang momentkernelmachineforclinicaldataminingtoinformmedicaldecisionmaking |