Novel AI Combining CNN and SVM to Predict Colorectal Cancer Prognosis and Mutational Signatures from HE Images.

Reducing recurrence following radical resection of colon cancer without over- or under-treatment remains a challenge. Postoperative adjuvant chemotherapy (Adj) is currently administered based solely on pathological tumor, node, and metastasis (TNM) stage. However, prognosis can vary significantly among patients with the same disease stage. Therefore, novel classification systems in addition to the TNM are necessary to inform decision-making regarding postoperative treatment strategies, especially stage II and III disease, and to minimize overtreatment and undertreatment with Adj. We developed a prognostic prediction system for colorectal cancer by using a combined convolutional neural network (CNN) and support vector machine (SVM) approach to extract features from hematoxyling and eosin staining (HE) images. We combined the TNM and our AI-based classification system into a TNM-AI (mTNM-AI) classification system with high discriminative power for recurrence-free survival (RFS). Furthermore, the cancer cell population recognized by this system as low risk of recurrence exhibited the mutational signature SBS87 as a genetic phenotype. The novel AI-based classification system developed here is expected to play an important role in prognostic prediction and personalized treatment selection in oncology.

Development of multiple AI pipelines that predict neoadjuvant chemotherapy response of breast cancer using H&E-stained tissues.

In recent years, the treatment of breast cancer has advanced dramatically and neoadjuvant chemotherapy (NAC) has become a common treatment method, especially for locally advanced breast cancer. However, other than the subtype of breast cancer, no clear factor indicating sensitivity to NAC has been identified. In this study, we attempted to use artificial intelligence (AI) to predict the effect of preoperative chemotherapy from hematoxylin and eosin images of pathological tissue obtained from needle biopsies prior to chemotherapy. Application of AI to pathological images typically uses a single machine-learning model such as support vector machines (SVMs) or deep convolutional neural networks (CNNs). However, cancer tissues are extremely diverse and learning with a realistic number of cases limits the prediction accuracy of a single model. In this study, we propose a novel pipeline system that uses three independent models each focusing on different characteristics of cancer atypia. Our system uses a CNN model to learn structural atypia from image patches and SVM and random forest models to learn nuclear atypia from fine-grained nuclear features extracted by image analysis methods. It was able to predict the NAC response with 95.15% accuracy on a test set of 103 unseen cases. We believe that this AI pipeline system will contribute to the adoption of personalized medicine in NAC therapy for breast cancer.

Predicting the prognosis of lower rectal cancer using preoperative magnetic resonance imaging with artificial intelligence.

BACKGROUND: There are various preoperative treatments that are useful for controlling local or distant metastases in lower rectal cancer. For planning perioperative management, preoperative stratification of optimal treatment strategies for each case is required. However, a stratification method has not yet been established. Therefore, we attempted to predict the prognosis of lower rectal cancer using preoperative magnetic resonance imaging (MRI) with artificial intelligence (AI). METHODS: This study included 54 patients [male:female ratio was 37:17, median age 70 years (range 49-107 years)] with lower rectal cancer who could be curatively resected without preoperative treatment at Tokyo Medical University Hospital from January 2010 to February 2017. In total, 878 preoperative T2 MRIs were analyzed. The primary endpoint was the presence or absence of recurrence, which was evaluated using the area under the receiver operating characteristic curve. The secondary endpoint was recurrence-free survival (RFS), which was evaluated using the Kaplan-Meier curve of the predicted recurrence (AI stage 1) and predicted recurrence-free (AI stage 0) groups. RESULTS: For recurrence prediction, the area under the curve (AUC) values for learning and test cases were 0.748 and 0.757, respectively. For prediction of recurrence in each case, the AUC values were 0.740 and 0.875, respectively. The 5-year RFS rates, according to the postoperative pathologic stage for all patients, were 100%, 64%, and 50% for stages 1, 2, and 3, respectively (p = 0.107). The 5-year RFS rates for AI stages 0 and 1 were 97% and 10%, respectively (p < 0.001 significant difference). CONCLUSIONS: We developed a prognostic model using AI and preoperative MRI images of patients with lower rectal cancer who had not undergone preoperative treatment, and the model could be useful in comparison with pathological classification.

Noise-driven growth rate gain in clonal cellular populations.

Cellular populations in both nature and the laboratory are composed of phenotypically heterogeneous individuals that compete with each other resulting in complex population dynamics. Predicting population growth characteristics based on knowledge of heterogeneous single-cell dynamics remains challenging. By observing groups of cells for hundreds of generations at single-cell resolution, we reveal that growth noise causes clonal populations of Escherichia coli to double faster than the mean doubling time of their constituent single cells across a broad set of balanced-growth conditions. We show that the population-level growth rate gain as well as age structures of populations and of cell lineages in competition are predictable. Furthermore, we theoretically reveal that the growth rate gain can be linked with the relative entropy of lineage generation time distributions. Unexpectedly, we find an empirical linear relation between the means and the variances of generation times across conditions, which provides a general constraint on maximal growth rates. Together, these results demonstrate a fundamental benefit of noise for population growth, and identify a growth law that sets a "speed limit" for proliferation.