A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks
Sperm identification and selection is an essential task when processing human testicular samples for in vitro fertilization. Locating and identifying sperm cell(s) in human testicular biopsy samples is labor intensive and time consuming. We developed a new computer-aided sperm analysis (CASA) system...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wolters Kluwer Medknow Publications
2021-01-01
|
Series: | Asian Journal of Andrology |
Subjects: | |
Online Access: | http://www.ajandrology.com/article.asp?issn=1008-682X;year=2021;volume=23;issue=2;spage=135;epage=139;aulast= |
id |
doaj-57e62c00a6854f0ca1bd60ae5da7c099 |
---|---|
record_format |
Article |
spelling |
doaj-57e62c00a6854f0ca1bd60ae5da7c0992021-03-29T10:44:39ZengWolters Kluwer Medknow PublicationsAsian Journal of Andrology1008-682X1745-72622021-01-0123213513910.4103/aja.aja_66_20A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networksDaniel J WuOdgerel BadamjavVikrant V ReddyMichael EisenbergBarry BehrSperm identification and selection is an essential task when processing human testicular samples for in vitro fertilization. Locating and identifying sperm cell(s) in human testicular biopsy samples is labor intensive and time consuming. We developed a new computer-aided sperm analysis (CASA) system, which utilizes deep learning for near human-level performance on testicular sperm extraction (TESE), trained on a custom dataset. The system automates the identification of sperm in testicular biopsy samples. A dataset of 702 de-identified images from testicular biopsy samples of 30 patients was collected. Each image was normalized and passed through glare filters and diffraction correction. The data were split 80%, 10%, and 10% into training, validation, and test sets, respectively. Then, a deep object detection network, composed of a feature extraction network and object detection network, was trained on this dataset. The model was benchmarked against embryologists' performance on the detection task. Our deep learning CASA system achieved a mean average precision (mAP) of 0.741, with an average recall (AR) of 0.376 on our dataset. Our proposed method can work in real time; its speed is effectively limited only by the imaging speed of the microscope. Our results indicate that deep learning-based technologies can improve the efficiency of finding sperm in testicular biopsy samples.http://www.ajandrology.com/article.asp?issn=1008-682X;year=2021;volume=23;issue=2;spage=135;epage=139;aulast=artificial intelligence; computer vision; male infertility; microdissection testicular sperm extraction; sperm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Daniel J Wu Odgerel Badamjav Vikrant V Reddy Michael Eisenberg Barry Behr |
spellingShingle |
Daniel J Wu Odgerel Badamjav Vikrant V Reddy Michael Eisenberg Barry Behr A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks Asian Journal of Andrology artificial intelligence; computer vision; male infertility; microdissection testicular sperm extraction; sperm |
author_facet |
Daniel J Wu Odgerel Badamjav Vikrant V Reddy Michael Eisenberg Barry Behr |
author_sort |
Daniel J Wu |
title |
A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
title_short |
A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
title_full |
A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
title_fullStr |
A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
title_full_unstemmed |
A preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
title_sort |
preliminary study of sperm identification in microdissection testicular sperm extraction samples with deep convolutional neural networks |
publisher |
Wolters Kluwer Medknow Publications |
series |
Asian Journal of Andrology |
issn |
1008-682X 1745-7262 |
publishDate |
2021-01-01 |
description |
Sperm identification and selection is an essential task when processing human testicular samples for in vitro fertilization. Locating and identifying sperm cell(s) in human testicular biopsy samples is labor intensive and time consuming. We developed a new computer-aided sperm analysis (CASA) system, which utilizes deep learning for near human-level performance on testicular sperm extraction (TESE), trained on a custom dataset. The system automates the identification of sperm in testicular biopsy samples. A dataset of 702 de-identified images from testicular biopsy samples of 30 patients was collected. Each image was normalized and passed through glare filters and diffraction correction. The data were split 80%, 10%, and 10% into training, validation, and test sets, respectively. Then, a deep object detection network, composed of a feature extraction network and object detection network, was trained on this dataset. The model was benchmarked against embryologists' performance on the detection task. Our deep learning CASA system achieved a mean average precision (mAP) of 0.741, with an average recall (AR) of 0.376 on our dataset. Our proposed method can work in real time; its speed is effectively limited only by the imaging speed of the microscope. Our results indicate that deep learning-based technologies can improve the efficiency of finding sperm in testicular biopsy samples. |
topic |
artificial intelligence; computer vision; male infertility; microdissection testicular sperm extraction; sperm |
url |
http://www.ajandrology.com/article.asp?issn=1008-682X;year=2021;volume=23;issue=2;spage=135;epage=139;aulast= |
work_keys_str_mv |
AT danieljwu apreliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT odgerelbadamjav apreliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT vikrantvreddy apreliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT michaeleisenberg apreliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT barrybehr apreliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT danieljwu preliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT odgerelbadamjav preliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT vikrantvreddy preliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT michaeleisenberg preliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks AT barrybehr preliminarystudyofspermidentificationinmicrodissectiontesticularspermextractionsampleswithdeepconvolutionalneuralnetworks |
_version_ |
1724198742263857152 |