Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients
Artificial intelligence (AI) is increasingly being used to improve patient care and management. In this paper, we propose explainable AI (XAI) models for predicting severe hypoglycemia (SH) and diabetic ketoacidosis (DKA) episodes in adults with type 1 diabetes (T1D) and relapses in adults with rela...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-01-01
|
| Series: | Informatics in Medicine Unlocked |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352914823001958 |
| _version_ | 1797646592062259200 |
|---|---|
| author | Anusha Ihalapathirana Konstantina Chalkou Pekka Siirtola Satu Tamminen Gunjan Chandra Pascal Benkert Jens Kuhle Georgia Salanti Juha Röning |
| author_facet | Anusha Ihalapathirana Konstantina Chalkou Pekka Siirtola Satu Tamminen Gunjan Chandra Pascal Benkert Jens Kuhle Georgia Salanti Juha Röning |
| author_sort | Anusha Ihalapathirana |
| collection | DOAJ |
| description | Artificial intelligence (AI) is increasingly being used to improve patient care and management. In this paper, we propose explainable AI (XAI) models for predicting severe hypoglycemia (SH) and diabetic ketoacidosis (DKA) episodes in adults with type 1 diabetes (T1D) and relapses in adults with relapsing-remitting multiple sclerosis (RRMS). We follow a three-step process in this study: (1) develop baseline machine learning (ML) models, (2) improve the models using ReliefF feature selection technique, and develop sex-stratified models, (3) explain the models and their results using SHapley Additive exPlanations (SHAP). We built six ML models (XGBoost, LightGBM, CatBoost, AdaBoost, random forest, and linear regression) for all scenarios. Applying the ReliefF feature selection led to improved model performance in predicting all outcomes compared to the baseline models. Additionally, sex-stratified models further improved the prediction of SH episodes and relapses. The F1 scores for predicting SH episodes in male and female patients were 84.07% and 84.95%, respectively, and the DKA prediction model achieved an F1 score of 78.67%. The proposed relapse prediction models outperformed existing models with F1 scores of 84.55% (males) and 76.11% (females), and ROCs of 70.26% (males) and 69.05% (females). Our results highlight the importance of considering sex differences, socioeconomic factors, and physical and mental health in medical outcome prediction. Boosting ML algorithms were found to be effective in detecting SH and DKA in T1D patients and relapses in RRMS patients compared to conventional tree-based ML and statistical models. |
| first_indexed | 2024-03-11T15:04:49Z |
| format | Article |
| id | doaj.art-7f195375cdf64be5aa582f9f95e02d8e |
| institution | Directory Open Access Journal |
| issn | 2352-9148 |
| language | English |
| last_indexed | 2024-03-11T15:04:49Z |
| publishDate | 2023-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Informatics in Medicine Unlocked |
| spelling | doaj.art-7f195375cdf64be5aa582f9f95e02d8e2023-10-30T06:05:11ZengElsevierInformatics in Medicine Unlocked2352-91482023-01-0142101349Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patientsAnusha Ihalapathirana0Konstantina Chalkou1Pekka Siirtola2Satu Tamminen3Gunjan Chandra4Pascal Benkert5Jens Kuhle6Georgia Salanti7Juha Röning8Biomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, FI-90014, Finland; Corresponding author.Institute of Social and Preventive Medicine, University of Bern, Bern, CH-3012, SwitzerlandBiomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, FI-90014, FinlandBiomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, FI-90014, FinlandBiomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, FI-90014, FinlandClinical Trial Unit, Department of Clinical Research, University Hospital Basel, University of Basel, Basel, 4001, SwitzerlandMultiple Sclerosis Centre, Neurologic Clinic and Policlinic, Departments of Head, Spine and Neuromedicine, Biomedicine and Clinical Research, University Hospital Basel and University of Basel, Basel, 4001, Switzerland; Research Center for Clinical Neuroimmunology and Neuroscience (RC2NB), University Hospital and University of Basel, Basel, 4001, SwitzerlandInstitute of Social and Preventive Medicine, University of Bern, Bern, CH-3012, SwitzerlandBiomimetics and Intelligent Systems Group, Faculty of Information Technology and Electrical Engineering, University of Oulu, Oulu, FI-90014, FinlandArtificial intelligence (AI) is increasingly being used to improve patient care and management. In this paper, we propose explainable AI (XAI) models for predicting severe hypoglycemia (SH) and diabetic ketoacidosis (DKA) episodes in adults with type 1 diabetes (T1D) and relapses in adults with relapsing-remitting multiple sclerosis (RRMS). We follow a three-step process in this study: (1) develop baseline machine learning (ML) models, (2) improve the models using ReliefF feature selection technique, and develop sex-stratified models, (3) explain the models and their results using SHapley Additive exPlanations (SHAP). We built six ML models (XGBoost, LightGBM, CatBoost, AdaBoost, random forest, and linear regression) for all scenarios. Applying the ReliefF feature selection led to improved model performance in predicting all outcomes compared to the baseline models. Additionally, sex-stratified models further improved the prediction of SH episodes and relapses. The F1 scores for predicting SH episodes in male and female patients were 84.07% and 84.95%, respectively, and the DKA prediction model achieved an F1 score of 78.67%. The proposed relapse prediction models outperformed existing models with F1 scores of 84.55% (males) and 76.11% (females), and ROCs of 70.26% (males) and 69.05% (females). Our results highlight the importance of considering sex differences, socioeconomic factors, and physical and mental health in medical outcome prediction. Boosting ML algorithms were found to be effective in detecting SH and DKA in T1D patients and relapses in RRMS patients compared to conventional tree-based ML and statistical models.http://www.sciencedirect.com/science/article/pii/S2352914823001958Explainable AIT1DRRMSSevere hypoglycemiaDiabetic ketoacidosisRelapses |
| spellingShingle | Anusha Ihalapathirana Konstantina Chalkou Pekka Siirtola Satu Tamminen Gunjan Chandra Pascal Benkert Jens Kuhle Georgia Salanti Juha Röning Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients Informatics in Medicine Unlocked Explainable AI T1D RRMS Severe hypoglycemia Diabetic ketoacidosis Relapses |
| title | Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients |
| title_full | Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients |
| title_fullStr | Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients |
| title_full_unstemmed | Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients |
| title_short | Explainable Artificial Intelligence to predict clinical outcomes in type 1 diabetes and relapsing-remitting multiple sclerosis adult patients |
| title_sort | explainable artificial intelligence to predict clinical outcomes in type 1 diabetes and relapsing remitting multiple sclerosis adult patients |
| topic | Explainable AI T1D RRMS Severe hypoglycemia Diabetic ketoacidosis Relapses |
| url | http://www.sciencedirect.com/science/article/pii/S2352914823001958 |
| work_keys_str_mv | AT anushaihalapathirana explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT konstantinachalkou explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT pekkasiirtola explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT satutamminen explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT gunjanchandra explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT pascalbenkert explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT jenskuhle explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT georgiasalanti explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients AT juharoning explainableartificialintelligencetopredictclinicaloutcomesintype1diabetesandrelapsingremittingmultiplesclerosisadultpatients |