**5. Conclusions**

The required number of annotated cases for accurate organ segmentation with a deep learning network may be lower than expected. The marginal benefit of more data may diminish after reaching a threshold number of cases in a deep learning network. In this study of prostate organ segmentation, the U-Net CNN plateaued at 160 cases.

**Author Contributions:** Conceptualization, M.B., D.C., and P.C.; methodology, M.B., and R.H.; software, M.B., C.C., and P.C.; validation, M.B., C.C., and P.C.; formal analysis, M.B., C.C., and P.C.; investigation, M.B., R.H., D.C., and P.C.; resources, D.C., E.U., and P.C.; data curation, M.B., R.H., A.U., J.G.-B., E.U., and P.C.; writing—original

draft preparation, M.B.; writing—review and editing, M.B., R.H., A.U., J.G.-B., M.S., and D.C.; visualization, M.B., R.H., C.C., D.C., and P.C.; supervision, R.H., D.C., and P.C.; project administration, R.H., D.C., and P.C.; funding acquisition, M.B., D.C., and P.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research is supported by a Radiological Society of North America Medical Student Research Grant (RMS1902) and additionally by an Alpha Omega Alpha Carolyn L. Kuckein Student Research Fellowship.

**Disclosures:** Author Peter Chang, MD, is a cofounder and shareholder of Avicenna.ai, a medical imaging startup. Author Daniel Chow, MD, is a shareholder of Avicenna.ai, a medical imaging startup.

**Conflicts of Interest:** The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
