1
|
Ai D, Cui C, Tang Y, Wang Y, Zhang N, Zhang C, Zhen Y, Li G, Huang K, Liu G, Chen Z, Zhang W, Wu R. Machine learning model for predicting physical activity related bleeding risk in Chinese boys with haemophilia A. Thromb Res 2023; 232:43-53. [PMID: 37931538 DOI: 10.1016/j.thromres.2023.10.012] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/23/2023] [Revised: 10/11/2023] [Accepted: 10/20/2023] [Indexed: 11/08/2023]
Abstract
BACKGROUND Physical activity is a crucial part of an active lifestyle for haemophiliac children. However, the fear of bleeds has been identified as barriers to participating physical activity for haemophiliac children even with prophylaxis. Lack of evidence and metrics driven by data is key problem. OBJECTIVES We aim to develop machine learning models based on clinical data with multiple potential factors considered to predict risk of physical activity bleeding for haemophilia children with prophylaxis. METHODS From this cohort study, we collected information on 98 haemophiliac children with adequate prophylaxis (trough FVIII:C level > 1 %). The involved potential predictor variables include demographic information, treatment information, physical activity, joint evaluation, and pharmacokinetic parameters, etc. We applied CoxPH, Random Survival Forests (RSF) and DeepSurv to construct prediction models for the risk of bleeding during physical activities. All three survival analysis models were internally and externally validated. RESULTS A total of 98 patients were enrolled in this study. Their median age was 7.9 (5.5, 10.2) years. The CoxPH, RSF and DeepSurv models' discriminative and calibration abilities were all high, and the RSF model had the best performance (Internal validation: C-index, 0.7648 ± 0.0139; Brier Score, 0.1098 ± 0.0015; External validation: C-index, 0.7260 ± 0.0154; Brier Score, 0.0930 ± 0.0018). The prediction curves demonstrated that the developed RSF model can distinguish the risks well between bleeding and non-bleeding patients, as well as patients with different levels of physical activity. Meanwhile, the feature importance analysis confirmed that physical activity bleeding was deduced by comprehensive effects of various factors, and the importance of different factors on bleeding outcome is discrepant. CONCLUSIONS This study revealed from the mechanism that it is necessary to incorporate multiple factors to accurately predict physical activity related bleeding risk. In clinical practice, the designed machine learning models can provide guidance for children with haemophilia A to positively participate in physical activity.
Collapse
Affiliation(s)
- Di Ai
- Haemophilia Comprehensive Care Center, Hematology Center, Beijing Key Laboratory of Pediatric Hematology-Oncology, National Key Discipline of Pediatrics (Capital Medical University), Key Laboratory of Major Diseases in Children, Ministry of Education, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, 100045, China
| | - Chang Cui
- State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China
| | - Yongqiang Tang
- State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China.
| | - Yan Wang
- Department of Rehabilitation, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Ningning Zhang
- Department of Radiology, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Chenyang Zhang
- State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China
| | - Yingzi Zhen
- Haemophilia Comprehensive Care Center, Hematology Center, Beijing Key Laboratory of Pediatric Hematology-Oncology, National Key Discipline of Pediatrics (Capital Medical University), Key Laboratory of Major Diseases in Children, Ministry of Education, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, 100045, China
| | - Gang Li
- Hematologic Disease Laboratory, Beijing Pediatric Research Institute, Beijing Children's Hospital, Capital Medical University, Beijing, China
| | - Kun Huang
- Haemophilia Comprehensive Care Center, Hematology Center, Beijing Key Laboratory of Pediatric Hematology-Oncology, National Key Discipline of Pediatrics (Capital Medical University), Key Laboratory of Major Diseases in Children, Ministry of Education, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, 100045, China
| | - Guoqing Liu
- Haemophilia Comprehensive Care Center, Hematology Center, Beijing Key Laboratory of Pediatric Hematology-Oncology, National Key Discipline of Pediatrics (Capital Medical University), Key Laboratory of Major Diseases in Children, Ministry of Education, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, 100045, China
| | - Zhenping Chen
- Hematologic Disease Laboratory, Beijing Pediatric Research Institute, Beijing Children's Hospital, Capital Medical University, Beijing, China.
| | - Wensheng Zhang
- State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China.
| | - Runhui Wu
- Haemophilia Comprehensive Care Center, Hematology Center, Beijing Key Laboratory of Pediatric Hematology-Oncology, National Key Discipline of Pediatrics (Capital Medical University), Key Laboratory of Major Diseases in Children, Ministry of Education, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, 100045, China.
| |
Collapse
|
2
|
Matlary RED, Holme PA, Glosli H, Rueegg CS, Grydeland M. Comparison of free-living physical activity measurements between ActiGraph GT3X-BT and Fitbit Charge 3 in young people with haemophilia. Haemophilia 2022; 28:e172-e180. [PMID: 35830613 PMCID: PMC9796296 DOI: 10.1111/hae.14624] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/05/2022] [Revised: 06/23/2022] [Accepted: 06/24/2022] [Indexed: 01/01/2023]
Abstract
INTRODUCTION Measurement of physical activity (PA) using commercial activity trackers such as Fitbit devices has become increasingly popular, also for people with haemophilia (PWH). The accuracy of the Fitbit model Charge 3 has not yet been examined. AIMS To compare the Fitbit Charge 3 against the research-grade accelerometer ActiGraph GT3X-BT in measuring average daily steps and minutes spent in different PA intensities. METHODS Twenty-four young PWH wore a wrist-worn Fitbit Charge 3 and hip-worn ActiGraph GT3X-BT simultaneously for seven consecutive days in free-living conditions. Correlation of and differences between the devices for daily averages of PA parameters were assessed using Pearson's correlation coefficient and paired t-test, respectively. Agreement between devices was assessed using Bland-Altman plots. RESULTS Twenty participants (mean age 21.8) were included in the analyses. We found moderate to high correlations between Fitbit and ActiGraph measured daily averages for all PA variables, but statistically significant differences between devices for all variables except daily minutes of moderate PA. Fitbit overestimated average daily steps, minutes of light, vigorous and moderate-to-vigorous PA. Bland-Altman plots showed a measurement bias between devices for all parameters with increasing overestimation by the Fitbit for higher volumes of PA. CONCLUSION The Fitbit Charge 3 overestimated steps and minutes of light, moderate and moderate-to-vigorous PA as compared to the ActiGraph GT3X-BT, and this bias increased with PA volume. The Fitbit should therefore be used with caution in research, and we advise users of the device to be cognizant of this overestimation.
Collapse
Affiliation(s)
- Ruth Elise D. Matlary
- Department of HaematologyOslo University HospitalOsloNorway,Institute of Clinical MedicineUniversity of OsloOsloNorway
| | - Pål André Holme
- Department of HaematologyOslo University HospitalOsloNorway,Institute of Clinical MedicineUniversity of OsloOsloNorway
| | - Heidi Glosli
- Centre for Rare DisordersOslo University HospitalOsloNorway,Department of Paediatric ResearchOslo University HospitalOsloNorway
| | - Corina Silvia Rueegg
- Oslo Centre for Biostatistics and EpidemiologyOslo University HospitalOsloNorway
| | - May Grydeland
- Department of Physical PerformanceNorwegian School of Sport SciencesOsloNorway
| |
Collapse
|
3
|
Ai D, Huang K, Li G, Zhen Y, Wu X, Zhang N, Huo A, Chen Z, Wu R. Exploration of the minimum necessary FVIII level at different physical activity levels in pediatric patients with hemophilia A. Front Pediatr 2022; 10:1045070. [PMID: 36389359 PMCID: PMC9665406 DOI: 10.3389/fped.2022.1045070] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/15/2022] [Accepted: 10/13/2022] [Indexed: 11/05/2022] Open
Abstract
BACKGROUND Physical activity can increase joint stability and reduce the risk of injury in hemophilia patients. There is limited clinical data on target trough FVIII levels during physical activity in hemophilia A patients. Hence, this study aimed to explore the target trough FVIII level required to avoid bleeding during different physical activities in hemophilia A patients. METHODS Patients with severe or moderate hemophilia A, who underwent pharmacokinetics (PK) tests at our center were enrolled in this study. Physical activities and clinical information such as bleeding were recorded. The FVIII level during physical activity was calculated by the WAPPS-Hemo. RESULTS A total of 105 patients were enrolled in this study. A total of 373 physical activities were recorded, of which 57.6% (215/373) was low-risk activities and the remaining 42.4% (158/373) was medium-risk activities. Most common physical activities were bicycling (59.0%), swimming (43.8%), running (48.6%), and jumping rope (41.0%). The FVIII trough level of low-risk physical activity was 3.8 IU/dl (AUC = 0.781, p = 0.002) and moderate-risk physical activity was 7.7 IU/dl (AUC = 0.809, p < 0.001). FVIII trough levels [low-risk activities: 6.1 (3.1, 13.2) IU/dl vs. 7.7 (2.3, 10.5) IU/dl, moderate-risk activities: 9.6 (5.8, 16.9) IU/dl vs. 10.2 (5.5, 11.0) IU/dl] were not statistically different between the mild arthropathy group and the moderate-severe arthropathy group. Multiple bleeding risk tended to increase with physical activities classified as moderate-risk (OR [95% CI]: 3.815 [1.766-8.238], p = 0.001). CONCLUSION The minimum necessary FVIII level increased with higher risk physical activity, irrespective of arthropathy.
Collapse
Affiliation(s)
- Di Ai
- Hematology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Kun Huang
- Hematology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Gang Li
- Hematologic Disease Laboratory, Beijing Pediatric Research Institute, Beijing Children's Hospital, Capital Medical University, Beijing, China
| | - Yingzi Zhen
- Hematology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Xinyi Wu
- Hematology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Ningning Zhang
- Department of Radiology, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Aihua Huo
- Department of Radiology, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| | - Zhenping Chen
- Hematologic Disease Laboratory, Beijing Pediatric Research Institute, Beijing Children's Hospital, Capital Medical University, Beijing, China
| | - Runhui Wu
- Hematology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China
| |
Collapse
|