Artificial intelligence to predict oncological outcome directly from hematoxylin and eosin-stained slides in urology

Deep learning to predict breast cancer sentinel lymph node status on INSEMA histological images

BACKGROUND

Sentinel lymph node (SLN) status is a clinically important prognostic biomarker in breast cancer and is used to guide therapy, especially for hormone receptor-positive, HER2-negative cases. However, invasive lymph node staging is increasingly omitted before therapy, and studies such as the randomised Intergroup Sentinel Mamma (INSEMA) trial address the potential for further de-escalation of axillary surgery. Therefore, it would be helpful to accurately predict the pretherapeutic sentinel status using medical images.

METHODS

Using a ResNet 50 architecture pretrained on ImageNet and a previously successful strategy, we trained deep learning (DL)-based image analysis algorithms to predict sentinel status on hematoxylin/eosin-stained images of predominantly luminal, primary breast tumours from the INSEMA trial and three additional, independent cohorts (The Cancer Genome Atlas (TCGA) and cohorts from the University hospitals of Mannheim and Regensburg), and compared their performance with that of a logistic regression using clinical data only. Performance on an INSEMA hold-out set was investigated in a blinded manner.

RESULTS

None of the generated image analysis algorithms yielded significantly better than random areas under the receiver operating characteristic curves on the test sets, including the hold-out test set from INSEMA. In contrast, the logistic regression fitted on the Mannheim cohort retained a better than random performance on INSEMA and Regensburg. Including the image analysis model output in the logistic regression did not improve performance further on INSEMA.

CONCLUSIONS

Employing DL-based image analysis on histological slides, we could not predict SLN status for unseen cases in the INSEMA trial and other predominantly luminal cohorts.

Impact of Tumor-Infiltrating Lymphocytes on Disease Progression in Human Papillomavirus-Related Oropharyngeal Carcinoma

OBJECTIVE

We aim to explore the prognostic value of tumor-infiltrating lymphocytes (TILs) in the primary tumor and metastatic lymph nodes of patients with HPV(+)OPSCC. We hypothesize that TILS density at both sites is associated with disease-free survival in HPV(+)OPSCC.

STUDY DESIGN

Matched case-control study among HPV(+)OPSCC patients who underwent intent-to-cure surgery. Cases developed locoregional or distant recurrence. Controls were matched based on age, sex, pathologic T, N, and overall stage, year of surgery, type of adjuvant treatment received, and the Adult Comorbidity Evaluation-27 (ACE-27) score.

SETTING

Single tertiary care center, May 2007 to December 2016.

METHODS

Tumoral TILs (tTILs) density was defined as % TILs; stromal TILs (sTILs) density was defined as absent/sparse or moderate/dense crowding. Associations between TILs and time to disease progression were assessed using Cox regression models.

RESULTS

Forty-four case-control pairs (N = 88) were included: 42 (48%) AJCC pStage I, 39 (44%) pStage II, and 7 (8%) pStage III. tTILs density ≥10% (hazard ratio [HR] 0.41, 95% confidence interval [CI] 0.17-0.99, p = .048) and a moderate/dense sTILs density (HR 0.21, 95% CI 0.06-0.75, p = .016) in the primary tumor were significantly associated with decreased risk of progression. TILs density in the lymph node was associated with decreased risk of progression but did not reach statistical significance. The tTILs and sTILs density correlated strongly between the primary tumor and lymph node. Concordance between the pathologists' was moderate (60%-70%).

CONCLUSIONS

In HPV(+)OPSCC, a higher density of tumoral and stromal TILs in the primary tumor and possibly the lymph node may predict a lower risk of disease progression.

A novel approach for automatic segmentation of prostate and its lesion regions on magnetic resonance imaging

Objective

To develop an accurate and automatic segmentation model based on convolution neural network to segment the prostate and its lesion regions.

Methods

Of all 180 subjects, 122 healthy individuals and 58 patients with prostate cancer were included. For each subject, all slices of the prostate were comprised in the DWIs. A novel DCNN is proposed to automatically segment the prostate and its lesion regions. This model is inspired by the U-Net model with the encoding-decoding path as the backbone, importing dense block, attention mechanism techniques, and group norm-Atrous Spatial Pyramidal Pooling. Data augmentation was used to avoid overfitting in training. In the experimental phase, the data set was randomly divided into a training (70%), testing set (30%). four-fold cross-validation methods were used to obtain results for each metric.

Results

The proposed model achieved in terms of Iou, Dice score, accuracy, sensitivity, 95% Hausdorff Distance, 86.82%,93.90%, 94.11%, 93.8%,7.84 for the prostate, 79.2%, 89.51%, 88.43%,89.31%,8.39 for lesion region in segmentation. Compared to the state-of-the-art models, FCN, U-Net, U-Net++, and ResU-Net, the segmentation model achieved more promising results.

Conclusion

The proposed model yielded excellent performance in accurate and automatic segmentation of the prostate and lesion regions, revealing that the novel deep convolutional neural network could be used in clinical disease treatment and diagnosis.