Knygos.lt nariams
Knygos.lt nariams
- Išsiųsime per 12–18 d.d.
/images/books/15574479/9789999328296.jpg)
- Autorius: Hamid Borkot Tulla
- Leidėjas: Eliva Press
- ISBN-10: 9999328296
- ISBN-13: 9789999328296
- Formatas: 15.2 x 22.9 x 0.5 cm, minkšti viršeliai
- Kalba: Anglų
Design and Development of a Medical Image Diagnosis System Based on Machine Learning (el. knyga) (skaityta knyga) | knygos.lt
Atsiliepimai
Aprašymas
"Design and Development of a Medical Image Diagnosis System Based on Machine Learning" by Md. Hamid Borkot Tulla is a pioneering undergraduate research project aimed at transforming breast cancer diagnosis. Leveraging the power of deep learning and transfer learning, this study deploys a fine-tuned ResNet50 convolutional neural network on the renowned BreaKHis dataset to classify histopathological breast tissue images as benign or malignant. The model achieved a remarkable accuracy of 81.28% and recall of 94.65%, providing reliable diagnostic support in clinical workflows. This research not only offers a practical AI-driven decision-support system for pathologists but also lays the groundwork for future multi-class classification models and real-time clinical integration in resource-constrained healthcare settings.
- Autorius: Hamid Borkot Tulla
- Leidėjas: Eliva Press
- ISBN-10: 9999328296
- ISBN-13: 9789999328296
- Formatas: 15.2 x 22.9 x 0.5 cm, minkšti viršeliai
- Kalba: Anglų
"Design and Development of a Medical Image Diagnosis System Based on Machine Learning" by Md. Hamid Borkot Tulla is a pioneering undergraduate research project aimed at transforming breast cancer diagnosis. Leveraging the power of deep learning and transfer learning, this study deploys a fine-tuned ResNet50 convolutional neural network on the renowned BreaKHis dataset to classify histopathological breast tissue images as benign or malignant. The model achieved a remarkable accuracy of 81.28% and recall of 94.65%, providing reliable diagnostic support in clinical workflows. This research not only offers a practical AI-driven decision-support system for pathologists but also lays the groundwork for future multi-class classification models and real-time clinical integration in resource-constrained healthcare settings.
Atsiliepimai