Fast Cell Parasites Detection with Neural Networks
Xu, X. F.; Talbot, S.; Lu, S. B.; Selvaraja, T.
Show abstract
Deep learning has dramatically improved the performance in many application areas such as image classification, object detection, speech recognition, drug discovery and etc since 2012. Where deep learning algorithms promise to discover the intricate hidden information inside the data by leveraging the large dataset, advanced model and computing power. Although deep learning techniques show medical expert level performance in a lot of medical applications, but some of the applications are still not explored or under explored due to the variation of the species. In this work, we studied the bright field based cell level Cryptosporidium and Giardia detection in the drink water with deep learning. Our experimental demonstrates that the new developed deep learning-based algorithm surpassed the handcrafted SVM based algorithm with above 97 percentage in accuracy and 700+fps in speed on embedded Jetson TX2 platform. Our research will lead to real-time and high accuracy label-free cell level Cryptosporidium and Giardia detection system in the future.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- A hybrid CNN-Random Forest algorithm for bacterial spore segmentation and classification in TEM images 95%
- Estimation of Three-Dimensional Chromatin Morphology for Nuclear Classification and Characterisation 95%
- A Convolution Based Computational Approach Towards DNA N6-methyladenine Site Identification and Motif Extraction in Rice Genome 94%
Similar papers in this journal
- An automated light trap to monitor moths (Lepidoptera) using computer vision-based tracking and deep learning 94%
- High-Throughput and Accurate 3D Scanning of Cattle Using Time-of-Flight Sensors and Deep Learning 93%
- A computational study on the role of parameters for identification of thyroid nodules by infrared images (and its comparison with real data) 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.