Tumor-infiltrating lymphocytes in breast FNA biopsy cytology: a predictor of tumor-infiltrating lymphocytes in histologic evaluation

SAMS-Net: Fusion of attention mechanism and multi-scale features network for tumor infiltrating lymphocytes segmentation

Automatic segmentation of tumor-infiltrating lymphocytes (TILs) from pathological images is essential for the prognosis and treatment of cancer. Deep learning technology has achieved great success in the segmentation task. It is still a challenge to realize accurate segmentation of TILs due to the phenomenon of blurred edges and adhesion of cells. To alleviate these problems, a squeeze-and-attention and multi-scale feature fusion network (SAMS-Net) based on codec structure, namely SAMS-Net, is proposed for the segmentation of TILs. Specifically, SAMS-Net utilizes the squeeze-and-attention module with the residual structure to fuse local and global context features and boost the spatial relevance of TILs images. Besides, a multi-scale feature fusion module is designed to capture TILs with large size differences by combining context information. The residual structure module integrates feature maps from different resolutions to strengthen the spatial resolution and offset the loss of spatial details. SAMS-Net is evaluated on the public TILs dataset and achieved dice similarity coefficient (DSC) of 87.2% and Intersection of Union (IoU) of 77.5%, which improved by 2.5% and 3.8% compared with UNet. These results demonstrate the great potential of SAMS-Net in TILs analysis and can further provide important evidence for the prognosis and treatment of cancer.

Investigating the role of core needle biopsy in evaluating tumor-stroma ratio (TSR) of invasive breast cancer: a retrospective study

PURPOSE

Tumor-stroma ratio (TSR) of invasive breast carcinoma has gained attention in recent years due to its prognostic significance. Previous studies showed TSR is a potential biomarker for indicating the tumor response to neoadjuvant chemotherapy. However, it is not clear how well TSR evaluation in biopsy specimens might reflect the TSR in resection specimens. We conducted a study to investigate whether biopsy evaluation of TSR can be an alternative method.

METHOD

We collected cases with invasive breast carcinoma of no special type (IBC-NST) from University of Yamanashi hospital between 2011 and 2017 whose biopsy and resection specimens both had a pathologically diagnosis of IBC-NST (n = 146). We conceptualized a method for evaluating TSR in biopsy specimens within a preliminary cohort (n = 50). Within the studied cohort (n = 96), biopsy-based TSR (b-TSR) and resection-based TSR (r-TSR) were scored by two pathologists. We then evaluated our method's validity and performance by measuring interobserver variability between the two pathologists, Spearman's correlation between b-TSR and r-TSR, and the receiver operating characteristics (ROC) analysis for defining stroma-rich and stroma-poor tumors.

RESULTS

Intra-class coefficient between the two pathologists was 0.59. The correlation coefficients between b-TSR and r-TSR in the two pathologists were 0.45 and 0.37. The ROC areas under the curve were 0.7 and 0.67. By considering an r-TSR of < 50% as stroma-rich, the sensitivity and specificity of detecting stroma-rich tumors were 64.1% and 66.7%, respectively, when b-TSR was < 40%.

CONCLUSION

Our current b-TSR evaluation method can provide information about r-TSR and facilitate pre-treatment therapy follow-up.