AI-powered Medical Sample Image Validation for Better Analysis and Diagnosis
Hardware: reComputer J1020v2 with NVIDIA Jetson Nano
Application: Histopathology Analysis
Industry: Smart Medical Examination
Deployment Location: United Kingdom
Solutions based on Jetson Edge AI are proven to be a preferred strategy for healthcare. By analyzing and processing large-scale images, and then analyzing their imaging quality, edge equipment can help medical experts quickly determine whether the microscope slide sample is defective or contaminated, and take timely rescue measures before the next research investment
Challenge
One prevalent challenge in healthcare systems is the susceptibility to time and resource leakage, manifesting as hidden inefficiencies that escalate significantly with scale.
Moreover, the accuracy of labeling tissue sample images is crucial, directly impacting operational efficiency and the reliability of AI model analysis and human diagnoses. While histopathological and radiological diagnoses’ images may be labeled manually, this method often leads to labeling errors. Consequently, the best recourse is to either repeat the diagnostic process or send the sample back to the laboratory for an update. However, this corrective action introduces delays.
Solution
To improve the analysis accuracy and keep data safe, the edge AI image processing quality validation system can be used widely, powered by the Jetson Nano edge device.
The system scans up to 80 tissue sample images in one circle, then all microscope sliding scans were segmented using a full convolutional neural network converted to TensorRT. Several error data qualities could be identified, such as areas outside the focus area in the slide, areas containing ink, tiling effects caused by stitching, areas of folding or tearing, and areas completely missed during scanning. This allows you to build a reliable data set that does not significantly affect the effectiveness of the entire training network because of any imperfect sample in the data set. It also gives doctors more information to make an accurate diagnosis.
Seeed NVIDIA Jetson Ecosystem
Seeed is an Elite partner for edge AI in the NVIDIA Partner Network. Explore more carrier boards, full system devices, customization services, use cases, and developer tools on Seeed’s NVIDIA Jetson ecosystem page.
Join the forefront of AI innovation with us! Harness the power of cutting-edge hardware and technology to revolutionize the deployment of machine learning in the real world across industries. Be a part of our mission to provide developers and enterprises with the best ML solutions available. Check out our success case study catalog to discover more edge AI possibilities!
Discover infinite computer vision application possibilities through our vision AI resource hub!
Take the first step and send us an email at edgeai@seeed.cc to become a part of this exciting journey!
Download our latest Jetson product Catalog to find one option that suits you well. If you can’t find the off-the-shelf Jetson hardware solution for your needs, please check out our customization services, and submit a new product inquiry to us at odm@seeed.cc for evaluation.
About Author
