Research Paper Volume 13, Issue 13 pp 17137—17154

Machine-intelligence for developing a potent signature to predict ovarian response to tailor assisted reproduction technology

Validation of CPLM and HPTM. (A) The calibration plot for training and validation data was used to analyze the consistency of poor ovarian response between the predicted value and the observed proportion in CPLM. (B) The calibration plot for training and validation data was used to analyze the consistency of poor ovarian response between the predicted value and the observed proportion in HPTM. (C) Correlation analysis of the CPLM score and relevant retrieved oocytes in the cohort. (D) Correlation analysis of the HPTM score and relevant retrieved oocytes in the cohort.

Figure 3. Validation of CPLM and HPTM. (A) The calibration plot for training and validation data was used to analyze the consistency of poor ovarian response between the predicted value and the observed proportion in CPLM. (B) The calibration plot for training and validation data was used to analyze the consistency of poor ovarian response between the predicted value and the observed proportion in HPTM. (C) Correlation analysis of the CPLM score and relevant retrieved oocytes in the cohort. (D) Correlation analysis of the HPTM score and relevant retrieved oocytes in the cohort.