JMIR Publications

ABSTRACT

Background:

The global decline in sperm concentration is a significant public health concern. Although standard semen analysis is widely conducted, it fails to identify functional sperm issues in many men with infertility. The sperm DNA Fragmentation Index (DFI) is a promising indicator of latent infertility; however, its clinical use is limited by high costs and complex procedures. Although image-based analytical methods, including artificial intelligence (AI), has the potential to estimate DFI from inexpensive microscopic images, the lack of high-quality multimodal datasets remains a major barrier.

Objective:

This paper describes the protocol for the Nippon Semen and Environmental Exposure Database (N-SEED) initiative. This initial phase aims to establish a standardized and scalable framework for prospective collection of biospecimens and multimodal data—including sperm microscopy image/video data, semen examination results, physical measurements, and questionnaire/interview data—from healthy adult men, and to statistically evaluate associations between sperm image data and sperm functional assessment measures (including DFI). These data will serve as a foundation for subsequent phases, including database expansion, refinement of image acquisition procedures, and stepwise development and external validation of image-based male fertility assessment methods (including AI-based approaches).

Methods:

This is a multi-institutional cross-sectional study with centralized clinical data collection at a single site in Japan to ensure high-quality standardization. Participants will be divided into two groups: Group 1 (n=25) focuses on the quality control of DFI measurement, and Group 2 (n=100) explores risk factors for male infertility in apparently healthy, non-clinic-based community volunteers. We will systematically collect microscopic sperm videos, DFI data, blood endocrine profiles, and urinary environmental chemical levels. Physical measurements and structured questionnaires regarding lifestyle, nutrition, and sexual function will also be administered. The primary outcome is the association between microscopic sperm image/video data and DFI.

Results:

The study protocol was approved by the relevant Institutional Review Boards on December 2025 and January 2026. Participant recruitment began on January 18, 2026, and recruitment and sample/data collection for the planned sample size were completed on February 21, 2026 (Group 1, n=25; Group 2, n=100). Biospecimen assays will be performed sequentially after collection and are planned to continue until March 2028. Primary association analyses are planned to be completed by March 2028.

Conclusions:

The N-SEED initiative establishes a standardized database framework that links sperm microscopy image/video data with DFI and extensive environmental and physiological indicators. This initial phase is expected to provide foundational infrastructure for future image-based analytical research, including stepwise development and external validation of noninvasive, low-cost screening approaches, rather than an immediately deployable clinical tool. Clinical Trial: UMIN Clinical Trials Registry UMIN000060395; https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000069079.