Human African Trypanosomiasis, commonly known as sleeping sickness, is still one of the NTDs that test the people of sub-Saharan Africa and the Latin American plateau. It has been marked by the World Health Organization (WHO) as one of those fatal diseases that should be diagnosed early. Therefore, it is necessary to develop efficient screening tools that will help solve the problem. This further led to the creation of Tryp dataset, which is the largest in its class. These are microscopy images taken from blood samples of people infected by Trypanosoma brucei brucei. After this, the approaches used to design Faster R-CNN, RetinaNet and YOLOv7 with regards to the Tryp dataset to correctly identify the images and to detect the parasites present within it were used. Out of these, YOLOv7 gave the maximum F1-score of 72%. However, considering that new versions of the You Only Look Once (YOLO) algorithm were later introduced, we feel there is still scope for improvement. So, we have come up with an advanced method with the latest YOLOv10 model which does the early identification and screening of the trypanosome parasite. A 23% gain on the mean average precision and a 4% gain on the overall F1-score was obtained by tuning the hyperparameters and fine-tuning the YOLOv10 model for the Tryp dataset compared to the previous best YOLOv7 model on the test partition of the dataset. This improved performance would point toward the capability of new object detection models, like YOLOv10, to better tackle NTD challenges, which in turn would result in better health outcomes for the affected communities.