Diabetes is one of the fatal diseases that play a vital role in the growth of other diseases in the human body. Controlling and curing diabetes in its early stages is the most significant technique to avoid its effects of diabetes. However, lack of awareness and expensive clinical tests are the primary reasons to skip clinical diagnosis and take preventive methods in lower-income countries like Bangladesh, Pakistan, and India. From this perspective, the study aims to build an automated machine learning (ML) model, which will predict diabetes at an early stage using socio-demographic characteristics rather than clinical attributes. Because clinical features are not always known to all people from lower-income countries. To find the best fit supervised ML classifier of the model, we applied six classification algorithms and found that RF outperformed with an accuracy of 99.36%. In addition, the most significant risk factors were found based on the SHAP value by all the applied classifiers. The study reveals that polyuria, polydipsia, and delayed healing are the most significant risk factors for growing diabetes. The findings indicate that the proposed model is highly capable of predicting diabetes in the early stages.

A cross-sectional study was carried out to determine the prevalence of subclinical mastitis (SCM) among medium to large scale household dairy farms in southwestern district, Jhenaidah, Bangladesh during July to December 2019. A total of 78 (n=100) lactating cows from household dairy farms (N=32) having three or more dairy cows were selected randomly as sampled populations. Milk samples were screened for SCM by using Surf Field Mastitis Test (SFMT). The prevalence of SCM varied among farm level [71.9% (95% CI: 53.3-86.3)], individual animal level [67.9% (95% CI: 56.4-86.3)] and quarter level [29.5% (95% CI: 24.5-34.9)]. Descriptive statistics represented the farmers and farm demography, characteristics of the sampled population, and overall management feature. Random Effect Logistic Regression identified, Body Condition Score (BCS) [OR=3.8 and 2.9, at cows level and quarter level respectively (BCS-2 vs. BCS-≥3)], and breed [OR=5.1 and 2.9, at cows level and quarter level respectively (HF× Sahiwal vs. HF × Local)] as potential risk factors. This study shows that SCM is highly prevalent in the study area, which is a major threat to the dairy industry's production performance. Regular screening by SFMT, proper hygiene, improve the management system, and farmer’s awareness is required to control the disease.