Prediction of Breast Cancer Recurrence Using a Deep Convolutional Neural Network Without Region-of-Interest Labeling

Harnessing artificial intelligence for predicting breast cancer recurrence: a systematic review of clinical and imaging data

Breast cancer is a leading cause of mortality among women, with recurrence prediction remaining a significant challenge. In this context, artificial intelligence application and its resources can serve as a powerful tool in analyzing large amounts of data and predicting cancer recurrence, potentially enabling personalized medical treatment and improving the patient's quality of life. Thus, the systematic review examines the role of AI in predicting breast cancer recurrence using clinical data, imaging data, and combined datasets. Support Vector Machine (SVM) and Neural Networks, especially when applied to combined data, demonstrate strong potential in improving prediction accuracy. SVMs are effective with high-dimensional clinical data, while Neural Networks in genetic and molecular analysis. Despite these advancements, limitations such as dataset diversity, sample size, and evaluation standardization persist, emphasizing the need for further research. AI integration in recurrence prediction offers promising prospects for personalized care but requires rigorous validation for safe clinical application.

Artificial intelligence in digital histopathology for predicting patient prognosis and treatment efficacy in breast cancer

INTRODUCTION

Histological images contain phenotypic information predictive of patient outcomes. Due to the heavy workload of pathologists, the time-consuming nature of quantitatively assessing histological features, and human eye limitations to recognize spatial patterns, manually extracting prognostic information in routine pathological workflows remains challenging. Digital pathology has facilitated the mining and quantification of these features utilizing whole-slide image (WSI) scanners and artificial intelligence (AI) algorithms. AI algorithms to identify image-based biomarkers from the tumor microenvironment (TME) have the potential to revolutionize the field of oncology, reducing delays between diagnosis and prognosis determination, allowing for rapid stratification of patients and prescription of optimal treatment regimes, thereby improving patient outcomes.

AREAS COVERED

In this review, the authors discuss how AI algorithms and digital pathology can predict breast cancer patient prognosis and treatment outcomes using image-based biomarkers, along with the challenges of adopting this technology in clinical settings.

EXPERT OPINION

The integration of AI and digital pathology presents significant potential for analyzing the TME and its diagnostic, prognostic, and predictive value in breast cancer patients. Widespread clinical adoption of AI faces ethical, regulatory, and technical challenges, although prospective trials may offer reassurance and promote uptake, ultimately improving patient outcomes by reducing diagnosis-to-prognosis delivery delays.

Nomogram prediction of the 70-gene signature (MammaPrint) binary and quartile categorized risk using medical history, imaging features and clinicopathological data among Chinese breast cancer patients

BACKGROUND

The 70-gene signature (70-GS, MammaPrint) test has been recommended by the main guidelines to evaluate prognosis and chemotherapy benefit of hormonal receptor positive human epidermal receptor 2 negative (HR + /Her2-) early breast cancer (BC). However, this expensive assay is not always accessible and affordable worldwide. Based on our previous study, we established nomogram models to predict the binary and quartile categorized risk of 70-GS.

METHODS

We retrospectively analyzed a consecutive cohort of 150 female patients with HR + /Her2- BC and eligible 70-GS test. Comparison of 40 parameters including the patients' medical history risk factors, imaging features and clinicopathological characteristics was performed between patients with high risk (N = 62) and low risk (N = 88) of 70-GS test, whereas risk calculations from established models including Clinical Treatment Score Post-5 years (CTS5), Immunohistochemistry 3 (IHC3) and Nottingham Prognostic Index (NPI) were also compared between high vs low binary risk of 70-GS and among ultra-high (N = 12), high (N = 50), low (N = 65) and ultra-low (N = 23) quartile categorized risk of 70-GS. The data of 150 patients were randomly split by 4:1 ratio with training set of 120 patients and testing set 30 patients. Univariate analyses and multivariate logistic regression were performed to establish the two nomogram models to predict the the binary and quartile categorized risk of 70-GS.

RESULTS

Compared to 70-GS low-risk patients, the high-risk patients had significantly less cardiovascular co-morbidity (p = 0.034), more grade 3 BC (p = 0.006), lower progesterone receptor (PR) positive percentage (p = 0.007), more Ki67 high BC (≥ 20%, p < 0.001) and no significant differences in all the imaging parameters of ultrasound and mammogram. The IHC3 risk and the NPI calculated score significantly correlated with both the binary and quartile categorized 70-GS risk classifications (both p < 0.001). The area under curve (AUC) of receiver-operating curve (ROC) of nomogram for binary risk prediction were 0.826 (C-index 0.903, 0.799-1.000) for training and 0.737 (C-index 0.785, 0.700-0.870) for validation dataset respectively. The AUC of ROC of nomogram for quartile risk prediction was 0.870 (C-index 0.854, 0.746-0.962) for training and 0.592 (C-index 0.769, 0.703-0.835) for testing set. The prediction accuracy of the nomogram for quartile categorized risk groups were 55.0% (likelihood ratio tests, p < 0.001) and 53.3% (p = 0.04) for training and validation, which more than double the baseline probability of 25%.

CONCLUSIONS

To our knowledge, we are the first to establish easy-to-use nomograms to predict the individualized binary (high vs low) and the quartile categorized (ultra-high, high, low and ultra-low) risk classification of 70-GS test with fair performance, which might provide information for treatment choice for those who have no access to the 70-GS testing.

Non-annotated renal histopathological image analysis with deep ensemble learning

BACKGROUND

Renal cancer is one of the leading causes of cancer-related deaths worldwide, and early detection of renal cancer can significantly improve the patients' survival rate. However, the manual analysis of renal tissue in the current clinical practices is labor-intensive, prone to inter-pathologist variations and easy to miss the important cancer markers, especially in the early stage.

METHODS

In this work, we developed deep convolutional neural network (CNN) based heterogeneous ensemble models for automated analysis of renal histopathological images without detailed annotations. The proposed method would first segment the histopathological tissue into patches with different magnification factors, then classify the generated patches into normal and tumor tissues using the pre-trained CNNs and lastly perform the deep ensemble learning to determine the final classification. The heterogeneous ensemble models consisted of CNN models from five deep learning architectures, namely VGG, ResNet, DenseNet, MobileNet, and EfficientNet. These CNN models were fine-tuned and used as base learners, they exhibited different performances and had great diversity in histopathological image analysis. The CNN models with superior classification accuracy (Acc) were then selected to undergo ensemble learning for the final classification. The performance of the investigated ensemble approaches was evaluated against the state-of-the-art literature.

RESULTS

The performance evaluation demonstrated the superiority of the proposed best performing ensembled model: five-CNN based weighted averaging model, with an Acc (99%), specificity (Sp) (98%), F1-score (F1) (99%) and area under the receiver operating characteristic (ROC) curve (98%) but slightly inferior recall (Re) (99%) compared to the literature.

CONCLUSIONS

The outstanding robustness of the developed ensemble model with a superiorly high-performance scores in the evaluated metrics suggested its reliability as a diagnosis system for assisting the pathologists in analyzing the renal histopathological tissues. It is expected that the proposed ensemble deep CNN models can greatly improve the early detection of renal cancer by making the diagnosis process more efficient, and less misdetection and misdiagnosis; subsequently, leading to higher patients' survival rate.

Evolution of Breast Cancer Recurrence Risk Prediction: A Systematic Review of Statistical and Machine Learning-Based Models

PURPOSE

Selection of appropriate adjuvant therapy to ultimately reduce the risk of breast cancer (BC) recurrence is a challenge for medical oncologists. Several automated risk prediction models have been developed using retrospective clinical data and have evolved significantly over the years in terms of predictors of recurrence, data usage, and predictive techniques (statistical/machine learning [ML]).

METHODS

Following PRISMA guidelines, we performed a systematic literature review of the aforementioned statistical and ML models published between January 2008 and December 2022 through searching five digital databases-PubMed, ScienceDirect, Scopus, Cochrane, and Web of Science. The comprehensive search yielded a total of 163 papers and after a screening process focusing on papers that dealt exclusively with statistical/ML methods, only 23 papers were deemed appropriate for further analysis. We benchmarked the studies on the basis of development, evaluation metrics, and validation strategy with an added emphasis on racial diversity of patients included in the studies.

RESULTS

In total, 30.4% of the included studies use statistical techniques, while 69.6% are ML-based. Among these, traditional ML models (support vector machines, decision tree, logistic regression, and naïve Bayes) are the most frequently used (26.1%) along with deep learning (26.1%). Deep learning and ensemble learning provide the most accurate predictions (AUC = 0.94 each).

CONCLUSION

ML-based prediction models exhibit outstanding performance, yet their practical applicability might be hindered by limited interpretability and reduced generalization. Moreover, predictive models for BC recurrence often focus on limited variables related to tumor, treatment, molecular, and clinical features. Imbalanced classes and the lack of open-source data sets impede model development and validation. Furthermore, existing models predominantly overlook African and Middle Eastern populations, as they are trained and validated mainly on Caucasian and Asian patients.

Exploring prognostic indicators in the pathological images of ovarian cancer based on a deep survival network

Background: Tumor pathology can assess patient prognosis based on a morphological deviation of tumor tissue from normal. Digitizing whole slide images (WSIs) of tissue enables the use of deep learning (DL) techniques in pathology, which may shed light on prognostic indicators of cancers, and avoid biases introduced by human experience. Purpose: We aim to explore new prognostic indicators of ovarian cancer (OC) patients using the DL framework on WSIs, and provide a valuable approach for OC risk stratification. Methods: We obtained the TCGA-OV dataset from the NIH Genomic Data Commons Data Portal database. The preprocessing of the dataset was comprised of three stages: 1) The WSIs and corresponding clinical data were paired and filtered based on a unique patient ID; 2) a weakly-supervised CLAM WSI-analysis tool was exploited to segment regions of interest; 3) the pre-trained model ResNet50 on ImageNet was employed to extract feature tensors. We proposed an attention-based network to predict a hazard score for each case. Furthermore, all cases were divided into a high-risk score group and a low-risk one according to the median as the threshold value. The multi-omics data of OC patients were used to assess the potential applications of the risk score. Finally, a nomogram based on risk scores and age features was established. Results: A total of 90 WSIs were processed, extracted, and fed into the attention-based network. The mean value of the resulting C-index was 0.5789 (0.5096-0.6053), and the resulting p-value was 0.00845. Moreover, the risk score showed a better prediction ability in the HRD + subgroup. Conclusion: Our deep learning framework is a promising method for searching WSIs, and providing a valuable clinical means for prognosis.

Uncertainty-informed deep learning models enable high-confidence predictions for digital histopathology

A model's ability to express its own predictive uncertainty is an essential attribute for maintaining clinical user confidence as computational biomarkers are deployed into real-world medical settings. In the domain of cancer digital histopathology, we describe a clinically-oriented approach to uncertainty quantification for whole-slide images, estimating uncertainty using dropout and calculating thresholds on training data to establish cutoffs for low- and high-confidence predictions. We train models to identify lung adenocarcinoma vs. squamous cell carcinoma and show that high-confidence predictions outperform predictions without uncertainty, in both cross-validation and testing on two large external datasets spanning multiple institutions. Our testing strategy closely approximates real-world application, with predictions generated on unsupervised, unannotated slides using predetermined thresholds. Furthermore, we show that uncertainty thresholding remains reliable in the setting of domain shift, with accurate high-confidence predictions of adenocarcinoma vs. squamous cell carcinoma for out-of-distribution, non-lung cancer cohorts.