ESTD PERPUSTAKAAN UNTAD
ESTD PERPUSTAKAAN UNTAD
| Judul IDENTIFIKASI KEONG ONCOMELANIA HUPENSIS LINDOENSIS SEBAGAI HOST SCHISTOSOMIASIS DENGAN MENGGUNAKAN CNN |
| Nama: MUH ALIF ALGHIFARI |
| Tahun: 2024 |
| Abstrak The World Health Organization reports schistosomiasis as a neglected tropical disease. In Indonesia, schistosomiasis is endemic in 3 regions of Central Sulawesi. In 2023, the prevalence rate of schistosomiasis in humans in Indonesia was 0.43%. Efforts are needed to achieve the government target of 0% prevalence in humans, snails, and mammals in 2025. Survey officers who do not recognize specific O. hupensis lindoensis snails need to ask officers who know. The identification system was made using the CNN (Convolutional Neural Network) algorithm with MobileNet architecture. With four classes and 1200 image data, the training accuracy is 93%, and the validation accuracy is 87%. The training loss function is 0.17, and the validation loss is 0.33. This research uses the Black Box testing method to test the functionality of the system with a result of 90% and Confusion Matrix testing precision with a result of 0.87. The results of this study can speed up, facilitate, and reduce the cost of snail prevalence surveys for officers and are useful for ordinary people to recognize this snail as the cause of schistosomiasis disease. |