RESUMEN
Background: Machine learning (ML) prepares and trains a model through supervised or unsupervised learning methods. Sputum, a respiratory tract secretion, is a common laboratory specimen that aids in diagnosing respiratory diseases, including pulmonary tuberculosis (TB). Gram stain is an easy, cost-effective stain, which may be applied to sputum smears to screen out an unsatisfactory sample. ML model may help in screening sputum smears. Methods: This collaborative project was carried out from June 2020-July 2021. In this study, a color-based segmentation ML algorithm using K-Means clustering was developed. A library of stained sputum smears was built. The Bartletts criteria (based on neutrophil and squamous cell count) for screening and selecting satisfactory sputum smears were used. A smartphone camera was used to take several photographs of satisfactory, as well as unsatisfactory, smears. The image segmentation algorithm was applied to medical image analysis, color-segmentation of sputum images was done. The hue saturation value (HSV) color ranges were defined on a prototype image. Then, all connected pixels were identified as a single object, and morphological operations were applied. Results: Usage of AI-driven model on the slide-image revealed the slide adequacy as the cell count was acceptable based on Bartlett's criteria. Both the manual cell counts (Range: 126-203 neutrophils, 14-47 squamous cells) and the model counts (Range: 117-242 neutrophils, 14-37 squamous cells) are within acceptable limits. Conclusion: The use of a model to screen a large number of sputum slides may be a boon in resource-limited settings where trained microscopists may not be easily available.
RESUMEN
The conventional paper-based system for malaria surveillance is time-consuming, difficult to track and resource-intensive. Few digital platforms are in use but wide-scale deployment and acceptability remain to be seen. To address this issue, we created a malaria surveillance mobile app that offers real-time data to stakeholders and establishes a centralised data repository. The MoSQuIT app was designed to collect data from the field and was integrated with a web-based platform for data integration and analysis. The MoSQuIT app was deployed on mobile phones of accredited social health activists (ASHA) working in international border villages in the northeast (NE) Indian states of Assam, Tripura and Arunachal Pradesh for 20 months in a phased manner. This paper shares the challenges and opportunities associated with the use of MoSQuIT for malaria surveillance. MoSQuIT employs the same data entry formats as the NVBDCP's malaria surveillance programme. Using this app, a total of 8221 fever cases were recorded, which included 1192 (14.5%) cases of P. falciparum malaria, 280 (3.4%) cases of P. vivax malaria and 52 (0.6%) mixed infection cases. Depending on network availability, GPS coordinates of the fever cases were acquired by the app. The present study demonstrated that mobile-phone-based malaria surveillance facilitates the quick transmission of data from the field to decision makers. Geospatial tagging of cases helped with easy visualisation of the case distribution for the identification of malaria-prone areas and potential outbreaks, especially in hilly and remote regions of Northeast India. However, to achieve the full operational potential of the system, operational challenges have to be overcome.