Publications
Publications
Development of machine learning backend medication adherence level classification, monitoring, and data collection tool in patients with type 2 diabetes in Ethiopia
Background: Medication adherence plays a crucial role in determining the health outcomes of patients, particularly those with chronic conditions like type 2 diabetes. Despite its significance, there is limited evidence regarding the use of machine learning (ML) algorithms to predict medication adherence within the Ethiopian population. The primary objective of this study was to develop and evaluate ML models designed to classify and monitor medication adherence levels among patients with type 2 diabetes in Ethiopia, to improve patient care and health outcomes.
Methods: Using a random sampling technique in a cross-sectional study, we obtained data from 403 patients with type 2 diabetes at the University of Gondar Comprehensive Specialized Hospital (UoGCSH), excluding 13 subjects who were unable to respond and 6 with incomplete data from an initial cohort of 422. Medication adherence was assessed using the General Medication Adherence Scale (GMAS), an eleven-item Likert scale questionnaire. The responses served as features to train and test machine learning (ML) models. To address data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied. The dataset was split using stratified K-fold cross-validation to preserve the distribution of adherence levels. Eight widely used ML algorithms were employed to develop the models, and their performance was evaluated using metrics such as accuracy, precision, recall, and F1 score. The best-performing model was subsequently deployed for further analysis.
Results: Out of 422 enrolled patients, 403 data samples were collected, with 11 features extracted from each respondent. To mitigate potential class imbalance, the dataset was increased to 620 samples using the Synthetic Minority Over-sampling Technique (SMOTE). Machine learning models including Logistic Regression (LR), Support Vector Machine (SVM), K Nearest Neighbor (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boost Classifier (GBC), Multilayer Perceptron (MLP), and 1D Convolutional Neural Network (1DCNN) were developed and evaluated. Although the performance differences among the models were subtle (within a range of 0.001), the SVM classifier outperformed the others, achieving a recall of 0.9979 and an AUC of 0.9998. Consequently, the SVM model was selected for deployment to monitor and detect patients’ medication adherence levels, enabling timely interventions to improve patient outcomes.
Conclusions: This study highlights a variety of machine learning (ML) models that can be effectively used to monitor and classify medication adherence in diabetic patients in Ethiopia. However, to fully realize the potential impact of digital health applications, further studies that include patients from diverse settings are necessary. Such research could enhance the generalizability of these models and provide insights into the broader applicability of digital tools for improving medication adherence and patient outcomes in varying healthcare contexts.
The application of machine learning approaches to classify and predict fertility rate in Ethiopia
Integrating machine learning (ML) models into healthcare systems is a rapidly evolving field with the potential to revolutionize care delivery. This study aimed to classify fertility rates and identify significant predictors using ML models among reproductive women in Ethiopia. This study utilized eight ML models in 5864 reproductive-age women using Ethiopian Demographic Health Survey (EDHS), 2019 data. Phyton programming language was used to develop these models. Predictors of fertility rate were determined using the feature important techniques. The performance of models was evaluated using accuracy, area under the curve (AUC), precision, recall, F1-score, specificity, and sensitivity. The mean age of participants was 32.7 (± 5.6) years. The random forest classifier (accuracy = 0.901 and AUC = 0.961) followed by a one-dimensional convolutional neural network (accuracy = 0.899 and AUC = 0.958), logistic regression (accuracy = 0.874 and AUC = 0.937), and gradient boost classifier (accuracy = 0.851 and AUC 0.927) were the top performing ML models. Family size, age, occupation, and education with an average importance score of 0.198, 0.151, 0.118, and 0.081, respectively were the top significant predictors of the fertility rate. The best ML models to classify and predict fertility rates were random forest, one-dimensional convolutional neural network, logistic regression, and gradient boost classifier. The findings on important factors of fertility rate can inform targeted public health, programs that address disparities related to family size, occupation, education, and other socioeconomic factors.
The impact of khat chewing on heart activity and rehabilitation therapy from khat addiction in healthy khat chewers
Khat is a flowering plant whose leaves and stems are chewed for excitement purposes in most of east African and Arabian countries. Khat can cause mood changes, increased alertness, hyperactivity, anxiety, elevated blood pressure, and heart diseases. However, the effect of khat on the heart has not been studied exclusively. The purpose of this study was to investigate the impact of khat chewing on heart activity and rehabilitation therapy from khat addiction in healthy khat chewers. ECG signals were recorded from 50 subjects (25 chewers and 25 controls) before and after chewing session to investigate the effect of khat on heart activity. In addition, ECG signals from 5 subjects were recorded on the first and eightieth day of rehabilitation therapy for investigating the effect of rehabilitation from khat addiction. All the collected signals were annotated, denoised and features were extracted and analysed. After chewing khat, the average heart rate of the chewers was increased by 5.85%, with 3 subjects out of 25 were prone to tachycardia. 1.66% QRS duration and 23.56% R-peak amplitude reduction were observed after chewing session. Moreover, heart rate variability was reduced by 19.74% indicating the effect of khat on suppressing sympathetic and parasympathetic nerve actions. After rehabilitation therapy, the average heart rate was reduced by 11.66%, while heart rate variability (HRV), QRS duration, and RR interval were increased by 25%, 3.49%, and 12.53%, respectively. Statistical analysis results also confirmed that there is a significance change (p < 0.05) in ECG feature among pre- and post-chewing session. Our findings demonstrate that, khat chewing raises heart rate, lowers heart rate variability, or puts the heart under stress by lowering R-peak amplitude and QRS duration, which in turn increases the risk of premature ventricular contraction and arrhythmia. The results also show that rehabilitation therapy from khat addiction has a major impact on restoring cardiac activity to normal levels.
A Framework for Locating Prescribed Medication at Pharmacies
Introduction: Accessibility of available medication at pharmacies is one of the core problems in the health sector of developing countries. The mechanism for optimally accessing the available drugs in pharmacies is unclear. Usually, patients in need are compelled to haphazardly switch between pharmacies in search of their prescription medications due to lack of information about the locations of pharmacies with required drug.
Objective: The primary objective of this study is to develop a framework that will simplify the process of identifying and locating nearest pharmacy when searching for prescribed medications.
Methods: Primary constraints (distance, drug cost, travel time, travel cost, opening and closing hours of pharmacies) in accessing required prescribed medications from pharmacies were identified from literature, and the client’s and pharmacies’ latitude and longitude coordinates were used to find the nearest pharmacies that have the required prescribed medication in stock.
Results: The framework with web application was developed and tested on simulated patients and pharmacies and was successful in optimizing the identified constraints.
Discussions: The framework will potentially reduce patient expenses and prevent delays in obtaining medication. It will also contribute for future pharmacy and e-Health information systems.
Muscle Fatigue Analysis and Stress Detection from Surface EMG and ECG Data Obtained Using Deep Learning for Upper-Limb Trauma Rehabilitation
Background: The repetitive nature of physical rehabilitation may result in excess muscular fatigue, which can adversely impact an individual's motor function, leading to discomfort or even physical injury. Moreover, individuals who have experienced trauma tend to encounter difficulties concentrating, which can significantly impede their physical capabilities. Regrettably, existing therapeutic approaches do not appear to consider the potential mental exhaustion of patients. This study aimed to create a bidirectional long short-term memory (Bi-LSTM) model for assessing muscle fatigue stage and mental stress conditions during physical rehabilitation of trauma-injured patients. Methods: Data corresponding to 188 EMG signals and 223 ECG signals were collected from the Jimma University physiotherapy clinic and prepared for signal processing. Since the 4th-order Butterworth filter performs better than the other filters, it was chosen to denoise the data. The data were then split at a ratio of 60:20:20 to train, validate, and test the data. Finally, the developed Bi-LSTM model was deployed. Results: The Bi-LSTM model achieved an accuracy of 95% for multiclass muscle fatigue classification, and 97% accuracy was achieved for the binary classification of mental stress. The GUI provides a setting appropriate for routine model usage. Conclusion: The results indicate that monitoring the muscle condition and mental status of traumatized patients can be performed in a clinical setting for effective physical rehabilitation.