Enhancing Leptospirosis Diagnosis through an integration of Machine Learning with Classification of Microscopic Agglutination Images

Leptospirosis, a widespread zoonotic disease, poses substantial challenges to global public health. In Malaysia, leptospirosis is an endemic disease, with annual cases peaking during the monsoon season. The microscopic agglutination test (MAT) is the gold-standard serological method for confirmation of leptospirosis. However, it is labour-intensive and time-consuming, as it relies on the subjective interpretation of medical lab technicians. This study investigates and describes the development of a semi-automated workflow for leptospira screening by integrating a TensorFlow and custom-designed Keras-based Deep Convolutional Neural Network (DCNN) with conventional MAT. We used a dataset of 442 positive and 442 negative MAT images, which consisted of a mixture of leptospira serovars to train the model. We then subjected our model to hyperparameter tuning, where we adjusted various settings to optimise the models performance. These settings included the number of convolutional layers, filters, kernel sizes, units in dense layers, activation functions, and the learning rate. We then tailored several convolutional layers to find the optimal balance between model complexity and performance. Verification of our tested model compared to the control samples (verified patient MAT results) achieved the following metrics: a Precision score of 0.8125, a Recall of 0.9286, and an F1-Score of 0.8667. Combining our model with the current Malaysia leptospira workflow can significantly speed up, reduce inaccuracies and improve the management of leptospirosis. Furthermore, the application of this model is practical and adaptable, making it suitable for other labs that observe MAT as their leptospira diagnosis. To our knowledge, this approach is Malaysias first hybrid diagnostic approach for leptospira diagnosis. Scaling up the dataset would enhance the models accuracy, making it adaptable in other regions where leptospirosis is endemic.