In This Article
- Clinical utility of traditional sperm testing is constrained by the need to deliver semen samples to a lab within one hour of production, labor-intensive sample analysis and the subjective nature interpreting results
- Data suggest a novel artificial intelligence-based system coupled with a smartphone-compatible optical hardware attachment mitigates constraints
- Advantages of automation include accuracy of results comparable to those of infertility laboratory technicians, the immediacy of results, lower cost and potential for remote or in-home use
Diagnosing fertility issues can be a challenge, but a multidisciplinary collaboration between clinicians in the Fertility Center at Massachusetts General Hospital and bioengineers at Brigham and Women's Hospital aims to change the field through a novel new application of machine learning, artificial intelligence and the latest smartphone technology.
The result is an automated sperm test that can easily be performed by a man using a smartphone fitted with an optical hardware device and an app that captures images of his sperm, which are then securely uploaded to the cloud for analysis by infertility specialists. This device and its capabilities are detailed in a recent PLoS ONE article.
Improving Access to Care
Fertility challenges can be quite common. Approximately 1/3 of infertility cases can be attributed to fertility issues in the male partner. While effective tests for analyzing semen for sperm viability are currently available, their clinical utility is often limited by technical, social and geographic factors.
"This new approach has the potential to change the way sperm testing is performed, meaning that testing could become more readily available and results will be accessible more quickly," says John Petrozza, MD, director of the Fertility Center at Mass General. "One of the key advantages here is point-of-care results."
"We're looking at it from a clinician's point of view and how we can improve patients' access to care," says Irene Dimitriadis, MD, a reproductive endocrinology and infertility fellow in the Fertility Center, and a collaborator on the project. "This definitely meets that goal."
Expanded Test Parameters
In 2017, Dr. Petrozza, embryologist Charles Bormann, PhD, and colleagues from the Division of Engineering in Medicine at Brigham and Women's Hospital published early results in Science Translational Medicine from a prototype validating their smartphone AI-based approach to the automated, remote analysis of semen. The initial prototype could analyze sperm morphology and motility.
Now, in response to enthusiasm for that early prototype, the research team has developed a more robust suite of test parameters to expand its clinical utility for infertility assessments.
The updated version of the device and app assesses:
- Sperm morphology and motility
- Hyaluronan binding assay (HBA) quantitative score describing sperm maturity and fertilization potential in a semen sample
- Sperm viability, which assesses sperm membrane integrity
- Sperm DNA fragmentation to assess the degree of DNA damage
Lowering Cost, Maintaining Quality
In addition to the convenience of remote and even in-home testing, the automated analysis maintains the quality of results of traditional lab tests. It also speeds analysis, which has the potential to greatly lower cost.
"We were able to actually get the same results as we're getting from using our very expensive laboratory equipment," says Dr. Dimitriadis. "Early results suggest that these tests may cost in the $5-$10 range."
Dr. Petrozza adds, "To be able to say the numbers are consistent whether you test the sperm sample using the phone app or bring it into the Mass General Fertility Center, to say you're going to get fairly robust identical ratings—that's amazing."
There are other advantages to this new approach according to the investigators. Making automated, remote sperm testing a reality not only expands infertility care, but also standardizes analyses to make it more objective from one sample to another.
According to Dr. Petrozza and Dr. Dimitriadis, it is plausible that within the next five years a commercial sperm testing kit could be as easily purchased over-the-counter as pregnancy test kits are now. Further, because the kits could be used anywhere with online access, they could help improve access to fertility care in underserved areas, from rural America to African villages.
The technology also has the potential to advance reproductive endocrinology research by providing large data sets of semen that can yield deeper insights into sperm pathology and its causes, the biology of conception and mechanisms involved in embryological anomalies.
Moving from Prototype to Product
The next steps for the research team include continuing to troubleshoot the technology, refining and validating usability and result quality, and working with funders to move the device to clinical trials.
"After collaborating with all the specialists on our team to successfully integrate AI and machine learning into health care, I'd say you would be hard-pressed to find another place that would have the collaborative spirit and expertise available to do this high level of work," says Dr. Petrozza. "I believe that, given our early results and experiences, this is a highly promising positive contribution to men's and family health."
Learn more about the Fertility Center
Refer a patient to the Fertility Center