1
|
Wang J, Jin Y, Jiang A, Chen W, Shan G, Gu Y, Ming Y, Li J, Yue C, Huang Z, Librach C, Lin G, Wang X, Zhao H, Sun Y, Zhang Z. Testing the generalizability and effectiveness of deep learning models among clinics: sperm detection as a pilot study. Reprod Biol Endocrinol 2024; 22:59. [PMID: 38778327 PMCID: PMC11110326 DOI: 10.1186/s12958-024-01232-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/03/2024] [Accepted: 05/14/2024] [Indexed: 05/25/2024] Open
Abstract
BACKGROUND Deep learning has been increasingly investigated for assisting clinical in vitro fertilization (IVF). The first technical step in many tasks is to visually detect and locate sperm, oocytes, and embryos in images. For clinical deployment of such deep learning models, different clinics use different image acquisition hardware and different sample preprocessing protocols, raising the concern over whether the reported accuracy of a deep learning model by one clinic could be reproduced in another clinic. Here we aim to investigate the effect of each imaging factor on the generalizability of object detection models, using sperm analysis as a pilot example. METHODS Ablation studies were performed using state-of-the-art models for detecting human sperm to quantitatively assess how model precision (false-positive detection) and recall (missed detection) were affected by imaging magnification, imaging mode, and sample preprocessing protocols. The results led to the hypothesis that the richness of image acquisition conditions in a training dataset deterministically affects model generalizability. The hypothesis was tested by first enriching the training dataset with a wide range of imaging conditions, then validated through internal blind tests on new samples and external multi-center clinical validations. RESULTS Ablation experiments revealed that removing subsets of data from the training dataset significantly reduced model precision. Removing raw sample images from the training dataset caused the largest drop in model precision, whereas removing 20x images caused the largest drop in model recall. by incorporating different imaging and sample preprocessing conditions into a rich training dataset, the model achieved an intraclass correlation coefficient (ICC) of 0.97 (95% CI: 0.94-0.99) for precision, and an ICC of 0.97 (95% CI: 0.93-0.99) for recall. Multi-center clinical validation showed no significant differences in model precision or recall across different clinics and applications. CONCLUSIONS The results validated the hypothesis that the richness of data in the training dataset is a key factor impacting model generalizability. These findings highlight the importance of diversity in a training dataset for model evaluation and suggest that future deep learning models in andrology and reproductive medicine should incorporate comprehensive feature sets for enhanced generalizability across clinics.
Collapse
Affiliation(s)
- Jiaqi Wang
- School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China
| | - Yufei Jin
- School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China
| | - Aojun Jiang
- Department of Mechanical Engineering, University of Toronto, Toronto, Canada
| | - Wenyuan Chen
- Department of Mechanical Engineering, University of Toronto, Toronto, Canada
| | - Guanqiao Shan
- Department of Mechanical Engineering, University of Toronto, Toronto, Canada
| | - Yifan Gu
- Institute of Reproductive and Stem Cell Engineering, School of Basic Medical Science, Central South University, Changsha, China
- Reproductive & Genetic Hospital of Citic-Xiangya, Changsha, China
| | - Yue Ming
- School of Medicine, The Chinese University of Hong Kong, Shenzhen, China
| | - Jichang Li
- School of Medicine, The Chinese University of Hong Kong, Shenzhen, China
| | - Chunfeng Yue
- Suzhou Boundless Medical Technology Ltd., Co., Suzhou, China
| | - Zongjie Huang
- Suzhou Boundless Medical Technology Ltd., Co., Suzhou, China
| | | | - Ge Lin
- Institute of Reproductive and Stem Cell Engineering, School of Basic Medical Science, Central South University, Changsha, China
- Reproductive & Genetic Hospital of Citic-Xiangya, Changsha, China
| | - Xibu Wang
- The 3rd Affiliated Hospital of Shenzhen University, Shenzhen, China
| | - Huan Zhao
- The 3rd Affiliated Hospital of Shenzhen University, Shenzhen, China.
| | - Yu Sun
- Department of Mechanical Engineering, University of Toronto, Toronto, Canada.
- Department of Computer Science, University of Toronto, Toronto, Canada.
- Institute of Biomedical Engineering, University of Toronto, Toronto, Canada.
- Department of Electrical and Computer Engineering, University of Toronto, Toronto, Canada.
| | - Zhuoran Zhang
- School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China.
| |
Collapse
|
2
|
Panner Selvam MK, Moharana AK, Baskaran S, Finelli R, Hudnall MC, Sikka SC. Current Updates on Involvement of Artificial Intelligence and Machine Learning in Semen Analysis. MEDICINA (KAUNAS, LITHUANIA) 2024; 60:279. [PMID: 38399566 PMCID: PMC10890589 DOI: 10.3390/medicina60020279] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/08/2024] [Revised: 01/29/2024] [Accepted: 01/31/2024] [Indexed: 02/25/2024]
Abstract
Background and Objectives: Infertility rates and the number of couples undergoing reproductive care have both increased substantially during the last few decades. Semen analysis is a crucial step in both the diagnosis and the treatment of male infertility. The accuracy of semen analysis results remains quite poor despite years of practice and advancements. Artificial intelligence (AI) algorithms, which can analyze and synthesize large amounts of data, can address the unique challenges involved in semen analysis due to the high objectivity of current methodologies. This review addresses recent AI advancements in semen analysis. Materials and Methods: A systematic literature search was performed in the PubMed database. Non-English articles and studies not related to humans were excluded. We extracted data related to AI algorithms or models used to evaluate semen parameters from the original studies, excluding abstracts, case reports, and meeting reports. Results: Of the 306 articles identified, 225 articles were rejected in the preliminary screening. The evaluation of the full texts of the remaining 81 publications resulted in the exclusion of another 48 articles, with a final inclusion of 33 original articles in this review. Conclusions: AI and machine learning are becoming increasingly popular in biomedical applications. The examination and selection of sperm by andrologists and embryologists may benefit greatly from using these algorithms. Furthermore, when bigger and more reliable datasets become accessible for training, these algorithms may improve over time.
Collapse
Affiliation(s)
- Manesh Kumar Panner Selvam
- Department of Urology, Tulane University School of Medicine, New Orleans, LA 70112, USA; (A.K.M.); (S.B.); (S.C.S.)
| | - Ajaya Kumar Moharana
- Department of Urology, Tulane University School of Medicine, New Orleans, LA 70112, USA; (A.K.M.); (S.B.); (S.C.S.)
- Redox Biology & Proteomics Laboratory, Department of Zoology, School of Life Sciences, Ravenshaw University, Cuttack 753003, Odisha, India
| | - Saradha Baskaran
- Department of Urology, Tulane University School of Medicine, New Orleans, LA 70112, USA; (A.K.M.); (S.B.); (S.C.S.)
| | | | | | - Suresh C. Sikka
- Department of Urology, Tulane University School of Medicine, New Orleans, LA 70112, USA; (A.K.M.); (S.B.); (S.C.S.)
| |
Collapse
|
3
|
Lustgarten Guahmich N, Borini E, Zaninovic N. Improving outcomes of assisted reproductive technologies using artificial intelligence for sperm selection. Fertil Steril 2023; 120:729-734. [PMID: 37307892 DOI: 10.1016/j.fertnstert.2023.06.009] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/06/2023] [Accepted: 06/06/2023] [Indexed: 06/14/2023]
Abstract
Within the field of assisted reproductive technology, artificial intelligence has become an attractive tool for potentially improving success rates. Recently, artificial intelligence-based tools for sperm evaluation and selection during intracytoplasmic sperm injection (ICSI) have been explored, mainly to improve fertilization outcomes and decrease variability within ICSI procedures. Although significant advances have been achieved in developing algorithms that track and rank single sperm in real-time during ICSI, the clinical benefits these might have in improving pregnancy rates from a single assisted reproductive technology cycle remain to be established.
Collapse
Affiliation(s)
- Nicole Lustgarten Guahmich
- Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medicine, New York, New York
| | - Elena Borini
- Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medicine, New York, New York
| | - Nikica Zaninovic
- Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medicine, New York, New York.
| |
Collapse
|
4
|
Si K, Huang B, Jin L. Application of artificial intelligence in gametes and embryos selection. HUM FERTIL 2023; 26:757-777. [PMID: 37705466 DOI: 10.1080/14647273.2023.2256980] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/28/2023] [Accepted: 07/22/2023] [Indexed: 09/15/2023]
Abstract
Gamete and embryo quality are critical to the success rate of Assisted Reproductive Technology (ART) cycles, but there remains a lack of methods to accurately measure the quality of sperm, oocytes and embryos. The ability of Artificial Intelligence (AI) technology to analyze large amounts of data, especially video and images, is particularly useful in gamete and embryo assessment and selection. The well-trained model has fast calculation speed and high accuracy, which can help embryologists to perform more objective gamete and embryo selection. Various artificial intelligence models have been developed for gamete and embryo assessment, some of which exhibit good performance. In this review, we summarize the latest applications of AI technology in semen analysis, as well as selection for sperm, oocyte and embryo, and discuss the existing problems and development directions of artificial intelligence in this field.
Collapse
Affiliation(s)
- Keyi Si
- Reproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and Technology, Wuhan, People's Republic of China
| | - Bo Huang
- Reproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and Technology, Wuhan, People's Republic of China
| | - Lei Jin
- Reproductive Medicine Center, Tongji Hospital, Tongji Medicine College, Huazhong University of Science and Technology, Wuhan, People's Republic of China
| |
Collapse
|