Blastocyst selection through an interpretable artificial intelligence method versus traditional morphology grading: study protocol for a randomised controlled trial
Bibliographic record
Abstract
INTRODUCTION: The quality of the blastocyst (day 5/6 embryo) selected for transfer is critical for the success of in vitro fertilisation (IVF) treatment. Embryologists perform blastocyst evaluation by observing the morphology of each blastocyst. Human assessment is subjective and inconsistent in predicting which blastocyst can result in a successful pregnancy or birth. Several artificial intelligence (AI) methods have been proposed to predict IVF outcomes from blastocyst images. However, the reasoning processes of these AI methods are uninterpretable, causing epistemic and ethical concerns that prevent their implementation in clinical practice. To address this issue, the authors developed a novel interpretable AI method for blastocyst selection. The method is clinically applicable because it is transparent to embryologists and allows them to understand its reasoning processes. This randomised controlled trial (RCT) aims to test the effectiveness of this blastocyst selection method with the aim of improving IVF outcomes. METHODS AND ANALYSIS: In this single-centre, single-blind RCT, we will enrol 1100 women aged 20-35 years undergoing their first cycle of IVF, with or without intracytoplasmic sperm injection. The study will be conducted at Nanjing Drum Tower Hospital, a public class A tertiary hospital in China. On the fifth day of embryo culture, participants with two or more usable blastocysts will be randomised in a 1:1 ratio to either the conventional morphology group or the AI group. The primary outcome is ongoing pregnancy, defined as a viable intrauterine pregnancy of 12 weeks gestation or more. ETHICS AND DISSEMINATION: The research ethics committee of the Nanjing Drum Tower Hospital approved this study (approval number: 2023-259-02). All participants will provide written informed consent prior to enrolment. The findings will be presented at international conferences and published in peer-reviewed journals. TRIAL REGISTRATION NUMBER: ChiCTR2300076851.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.036 | 0.043 |
| Meta-epidemiology (narrow) | 0.006 | 0.002 |
| Meta-epidemiology (broad) | 0.015 | 0.006 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.009 | 0.005 |
| Insufficient payload (model declined to judge) | 0.057 | 0.007 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".