Convolutional neural networks for segmentation and object detection of human semen

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

We compare a set of convolutional neural network (CNN) architectures for the task of segmenting and detecting human sperm cells in an image taken from a semen sample. In contrast to previous work, samples are not stained or washed to allow for full sperm quality analysis, making analysis harder due to clutter. Our results indicate that training on full images is superior to training on patches when class-skew is properly handled. Full image training including up-sampling during training proves to be beneficial in deep CNNs for pixel wise accuracy and detection performance. Predicted sperm cells are found by using connected components on the CNN predictions. We investigate optimization of a threshold parameter on the size of detected components. Our best network achieves 93.87% precision and 91.89% recall on our test dataset after thresholding outperforming a classical image analysis approach.

OriginalsprogEngelsk
TitelImage Analysis : 20th Scandinavian Conference, SCIA 2017, Tromsø, Norway, June 12–14, 2017, Proceedings, Part I
RedaktørerPuneet Sharma, Filippo Maria Bianchi
Antal sider10
Vol/bindPart 1
ForlagSpringer
Publikationsdato2017
Sider397-406
ISBN (Trykt)978-3-319-59125-4
ISBN (Elektronisk)978-3-319-59126-1
DOI
StatusUdgivet - 2017
Begivenhed20th Scandinavian Conference on Image Analysis - Tromsø, Norge
Varighed: 12 jun. 201714 jun. 2017
Konferencens nummer: 20

Konference

Konference20th Scandinavian Conference on Image Analysis
Nummer20
LandNorge
ByTromsø
Periode12/06/201714/06/2017
NavnLecture notes in computer science
Vol/bind10269
ISSN0302-9743

Links

ID: 184142886