npj Digital Medicine

After submission, three rounds of revision comments were given, one major revision and two minor revisions. The opinions of the two external reviewers were very positive. However, in the third round, they suddenly raised concerns about how to choose between fixed and random effects in the meta-analysis method. Following the Cochrane guidelines, the analysis was redone, and the results under both modes were consistent. But the other party did not accept it and chose to reject the submission. If that's the case, then all meta-analysis methods published by the journal are problematic and should be retracted. The editorial team is very unprofessional in handling this and took 6 months to reject it. Besides this methodological issue, they also claimed there were problems with novelty. If that was the case, why didn't they reject it in the first round? Why wait for 6 months and invite three rounds of revision? I really can't understand.

Supplement to the previous content:
2022.06.02 Accepted in principle but with some minor modifications allowed + submission of editable files.
2022.06.10 Submission.
2022.06.16 Official acceptance.

The work is about machine learning prediction of Alzheimer's disease in the elderly. There were three reviewers in the first round, with two providing many suggestions and raising some questions. One reviewer leaned towards rejection, stating that the title might be exaggerating the work. After adding two additional experiments, both reviewers agreed to accept the paper before the second round of review. The overall quality of the review was very high, and the suggestions made by the three reviewers were very pertinent, greatly improving the article itself.