Skip to content
DraconiticResearch
Search Sources API
190,869 studies31 sourcesUpdated 35 min ago
Peer-reviewedCC BY 4.0Journal of Clinical Medicine

Retrospective Analysis of Tuberculosis Treatment Duration and Outcomes Using Explainable Machine Learning on Malaysian Patient Data

  • Vimala Balakrishnan1,2
  • Jin Khye Tan1
  • Chin Lay Gan3,4
4 affiliations
  1. 1Faculty of Computer Science and Information Technology, Universiti Malaya, Kuala Lumpur 50603, Malaysia
  2. 2Department of Computer Science and Engineering, Korea University, Seoul 02841, Republic of Korea
  3. 3Centre for Management and Marketing Innovation, CoE for Business Innovation and Communication, Multimedia University, Persiaran Multimedia, Cyberjaya 63100, Selangor, Malaysia
  4. 4Faculty of Business, Multimedia University, Jalan Ayer Keroh Lama, Melaka 75450, Melaka, Malaysia
Read full textPDFDOICite

Abstract

Background/Objectives: This study utilizes machine learning (ML) models integrated with explainable artificial intelligence (XAI) to perform a retrospective analysis of factors associated with tuberculosis (TB) treatment duration and outcomes. Methods: Two datasets comprising completed TB patient records from a Malaysian hospital were analyzed. Significant features were identified using Pearson correlation for treatment duration and the Boruta algorithm for treatment outcome. Support Vector Regressor (SVR) was used to evaluate treatment duration, while Logistic Regression, Random Forest, and XGBoost were evaluated for treatment outcome classification. SHapley Additive Explanations (SHAP) was integrated to enhance model interpretability. Results: On the primary dataset (Dataset A), SVRPearson demonstrated the best performance for treatment duration (MAE = 44.18 days, R2 = 0.54), although performance decreased on the external dataset (Dataset B: MAE = 71.33 days, R2 = 0.11), partly due to missing feature imputations. For treatment outcome classification, XGBoostBoruta achieved an accuracy of 99.20% on Dataset A and 97.34% on Dataset B. SHAP feature attribution indicated that treatment outcome was the primary contributor to treatment duration, whereas treatment status and treatment duration were the dominant features contributing to the treatment outcome model. Conclusions: These findings demonstrate the value of explainable machine learning for characterizing factors associated with TB treatment duration and outcomes in completed treatment episodes and supporting retrospective clinical audit and post hoc case interpretation.

Related

  1. Extended treatment duration and recurrence in successfully treated drug-susceptible pulmonary tuberculosis with cavity on chest radiograph
    Chang-Seok Yoon, Tae-Ok Kim, Hong-Joon Shin +19
    PLOS OnePeer-reviewedCC BY 4.0Full textEndocrinology & metabolismEpidemiology & public healthPLOS
  2. Postoperative Antibiotic Duration and Treatment Failure After Aseptic Revision Arthroplasty With a Single Unexpected Positive Culture: A Retrospective Cohort Study.
    Hao-Chang Hsieh, Feng-Chih Kuo, Allen Herng Shouh Hsu +3
    Orthopaedic surgeryPeer-reviewedCC BYFull textSurgery & anaesthesiaEurope PMC
  3. Hypertension care cascade among adults in a Malaysian health surveillance cohort: a longitudinal evaluation of screening, diagnosis, treatment, and control
    Chiew Way Ang, Roshidi Ismail, Tin Tin Su
    BMC Public HealthPeer-reviewedCC BY-NC-ND 4.0Cardiovascular medicineCrossref
  4. Characteristics and Treatment Outcomes of Patients with Drug-Resistant Tuberculosis in Selected Facilities of O.R. Tambo District Municipality, Eastern Cape, South Africa: A Retrospective Cohort Study.
    Yanga Somjobo, Lindiwe Modest Faye, Mojisola Clara Hosu
    Epidemiologia (Basel, Switzerland)Peer-reviewedCC BYFull textInfectious diseaseEurope PMC
  5. Explainable machine learning for identifying insufficient sleep duration among preschool children in cold-region China: a cross-sectional study
    Xichao Zhang, Yuzi Diao, Yufei Liu +4
    BMC Public HealthPeer-reviewedCC BY-NC-ND 4.0Epidemiology & public healthCrossref

Cite

BibTeXRIS
BibTeX
@article{balakrishnan2026retrospective,
  title = {{Retrospective Analysis of Tuberculosis Treatment Duration and Outcomes Using Explainable Machine Learning on Malaysian Patient Data}},
  author = {Vimala Balakrishnan and Jin Khye Tan and Chin Lay Gan},
  journal = {Journal of Clinical Medicine},
  year = {2026},
  month = oct,
  doi = {10.3390/jcm15207888},
  url = {https://www.mdpi.com/2077-0383/15/20/7888},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Retrospective Analysis of Tuberculosis Treatment Duration and Outcomes Using Explainable Machine Learning on Malaysian Patient Data
AU  - Vimala Balakrishnan
AU  - Jin Khye Tan
AU  - Chin Lay Gan
PY  - 2026
DA  - 2026/10/01
JO  - Journal of Clinical Medicine
DO  - 10.3390/jcm15207888
UR  - https://www.mdpi.com/2077-0383/15/20/7888
AB  - Background/Objectives: This study utilizes machine learning (ML) models integrated with explainable artificial intelligence (XAI) to perform a retrospective analysis of factors associated with tuberculosis (TB) treatment duration and outcomes. Methods: Two datasets comprising completed TB patient records from a Malaysian hospital were analyzed. Significant features were identified using Pearson correlation for treatment duration and the Boruta algorithm for treatment outcome. Support Vector Regressor (SVR) was used to evaluate treatment duration, while Logistic Regression, Random Forest, and XGBoost were evaluated for treatment outcome classification. SHapley Additive Explanations (SHAP) was integrated to enhance model interpretability. Results: On the primary dataset (Dataset A), SVRPearson demonstrated the best performance for treatment duration (MAE = 44.18 days, R2 = 0.54), although performance decreased on the external dataset (Dataset B: MAE = 71.33 days, R2 = 0.11), partly due to missing feature imputations. For treatment outcome classification, XGBoostBoruta achieved an accuracy of 99.20% on Dataset A and 97.34% on Dataset B. SHAP feature attribution indicated that treatment outcome was the primary contributor to treatment duration, whereas treatment status and treatment duration were the dominant features contributing to the treatment outcome model. Conclusions: These findings demonstrate the value of explainable machine learning for characterizing factors associated with TB treatment duration and outcomes in completed treatment episodes and supporting retrospective clinical audit and post hoc case interpretation.
SN  - 2077-0383
ER  -