An AI Model Predicts the Efficacy of Neoadjuvant Chemotherapy for Breast Cancer: a Multicenter, Bidirectional Cohort Study

Neoadjuvant chemotherapy is an important part of the systematic treatment of breast cancer, and it is of great clinical significance to predict the efficacy of neoadjuvant chemotherapy in early stage. The emergence of multi-modal artificial intelligence model has brought new ideas for it. However, the limited ability of artificial intelligence to integrate multi-modal data, the lack of multi-modal models, and the insufficient level of evidence in clinical promotion of artificial intelligence are all scientific problems that need to be solved. In the early stage of the study, a variety of artificial intelligence accurate prediction and auxiliary diagnosis and treatment models for breast cancer were constructed based on magnetic resonance imaging and pathomics, etc., and the effectiveness of the models in predicting the curative effect of neoadjuvant chemotherapy for breast cancer was explored. In order to further improve the predictive efficiency of the model and fill the gap in the systematic study of multi-modal data fusion model, this clinical study intends to combine pathological images, magnetic resonance imaging, diagnostic report text and clinical variables to establish an artificial intelligence large language model based on multi-task and multi-modal data fusion to accurately predict the efficacy of neoadjuvant chemotherapy for breast cancer. A multicenter, bidirectional cohort study was conducted to explore the predictive effectiveness of the model.

Sex

Female

Inclusion Criteria:

* Women
* Pathological diagnosis of non-metastatic invasive breast cancer (stage II-III)
* At least 4 cycles of neoadjuvant chemotherapy
* Radical surgery was performed after neoadjuvant chemotherapy
* There are pathological images and reports of breast puncture specimens before neoadjuvant chemotherapy
* 'There are MRI images and reports of breast MRI within 2 weeks before neoadjuvant chemotherapy
* There are standard clinical records

Exclusion Criteria:

* Inflammatory breast cancer
* Bilateral breast cancer
* Newly diagnosed stage IV breast cancer
* Other tumors have not been completely removed or less than 3 years after surgery
* Treatment other than neoadjuvant therapy had been performed before surgery