5 January 2021 in The Journal of Pathology
10 November 2025 in Journal of Clinical Oncology
PURPOSE: Artificial intelligence (AI) holds significant promise for improving cancer diagnosis and treatment. Here, we present a foundation AI model for prognosis prediction on the basis of standard hematoxylin and eosin–stained histopathology slides. METHODS: In this multinational cohort study, we developed AI models to predict prognosis from histopathology images of patients with GI cancers. First, we trained a foundation model using over 130 million patches from 104,876 whole-slide images on the basis of self-supervised learning. Second, we fine-tuned deep learning models for predicting survival outcomes and validated them across seven cohorts, including 1,619 patients with gastric and esophageal cancers and 2,594 patients with colorectal cancer. We further assessed the model for predicting survival benefit from adjuvant chemotherapy. RESULTS: The AI models predicted disease-free survival and disease-specific survival with a concordance index of 0.726-0.797 for gastric cancer and 0.714-0.757 for colorectal cancer in the validation cohorts. The models stratified patients into high-risk and low-risk groups, with 5-year survival rates of 49%-52% versus 76%-92% in gastric cancer and 43%-72% versus 81%-98% in colorectal cancer. In multivariable analysis, the AI risk scores remained an independent prognostic factor after adjusting for clinicopathologic variables. Compared with stage alone, an integrated model consisting of stage and image information improved prognosis prediction across all validation cohorts. Finally, adjuvant chemotherapy was associated with improved survival in the high-risk group but not in the low-risk group (treatment-model interaction P = .01 and .006) for stage II/III gastric and colorectal cancer, respectively. CONCLUSION: The pathology foundation model can accurately predict survival outcomes and complement clinicopathologic factors in GI cancers. Pending prospective validation, it may be used to improve risk stratification and inform personalized adjuvant therapy.
1 November 2020 in JCO Clinical Cancer Informatics
PURPOSE: Deep learning (DL), a class of approaches involving self-learned discriminative features, is increasingly being applied to digital pathology (DP) images for tasks such as disease identification and segmentation of tissue primitives (eg, nuclei, glands, lymphocytes). One application of DP is in telepathology, which involves digitally transmitting DP slides over the Internet for secondary diagnosis by an expert at a remote location. Unfortunately, the places benefiting most from telepathology often have poor Internet quality, resulting in prohibitive transmission times of DP images. Image compression may help, but the degree to which image compression affects performance of DL algorithms has been largely unexplored. METHODS: We investigated the effects of image compression on the performance of DL strategies in the context of 3 representative use cases involving segmentation of nuclei (n = 137), segmentation of lymph node metastasis (n = 380), and lymphocyte detection (n = 100). For each use case, test images at various levels of compression (JPEG compression quality score ranging from 1-100 and JPEG2000 compression peak signal-to-noise ratio ranging from 18-100 dB) were evaluated by a DL classifier. Performance metrics including F1 score and area under the receiver operating characteristic curve were computed at the various compression levels. RESULTS: Our results suggest that DP images can be compressed by 85% while still maintaining the performance of the DL algorithms at 95% of what is achievable without any compression. Interestingly, the maximum compression level sustainable by DL algorithms is similar to where pathologists also reported difficulties in providing accurate interpretations. CONCLUSION: Our findings seem to suggest that in low-resource settings, DP images can be significantly compressed before transmission for DL-based telepathology applications.
5 March 2025 in IEEE Journal of Biomedical and Health Informatics
Applying deep learning to predict patient prognostic survival outcomes using histological whole-slide images (WSIs) and genomic data is challenging due to the morphological and transcriptomic heterogeneity present in the tumor microenvironment. Existing deep learning-enabled methods often exhibit learning biases, primarily because the genomic knowledge used to guide directional feature extraction from WSIs may be irrelevant or incomplete. This results in a suboptimal and sometimes myopic understanding of the overall pathological landscape, potentially overlooking crucial histological insights. To tackle these challenges, we propose the CounterFactual Bidirectional Co-Attention Transformer framework. By integrating a bidirectional co-attention layer, our framework fosters effective feature interactions between the genomic and histology modalities and ensures consistent identification of prognostic features from WSIs. Using counterfactual reasoning, our model utilizes causality to model unimodal and multimodal knowledge for cancer risk stratification. This approach directly addresses and reduces bias, enables the exploration of ’what-if' scenarios, and offers a deeper understanding of how different features influence survival outcomes. Our framework, validated across eight diverse cancer benchmark datasets from The Cancer Genome Atlas (TCGA), represents a major improvement over current histology-genomic model learning methods. It shows an average 2.5% improvement in c-index performance over 18 state-of-the-art models in predicting patient prognoses across eight cancer types. Our code is released at https://github.com/BusyJzy599/CFBCT-main.
1 September 1994 in Optics Letters
We consider the copropagation of three (pump, Stokes, and anti-Stokes) self-trapped beams that interact through the parametric four-photon-mixing process in a (1 + 1 dimension) uniform self-focusing nonlinear medium. The stationary propagation involving three trapped beams is shown to be possible below a critical frequency shift when the phase difference between the three waves is either matched or highly mismatched. Most importantly, the steady-state solutions are found to be stable, as in the case of presence of only one soliton beam, except that the two (in-phase) fundamental stationary states launched in parallel will not return to their original states periodically after collision; rather, they may repel each other on collision and propagate in symmetrically tilted directions thereafter with increasing separation.
5 April 2023 in Kidney360
Key Points: Computational image analysis allows for the extraction of new information from whole-slide images with potential clinical relevance. Peritubular capillary (PTC) density is decreased in areas of interstitial fibrosis and tubular atrophy when measured in interstitial fractional space. PTC shape (aspect ratio) is associated with clinical outcome in glomerular diseases. Background: The association between peritubular capillary (PTC) density and disease progression has been studied in a variety of kidney diseases using immunohistochemistry. However, other PTC attributes, such as PTC shape, have not been explored yet. The recent development of computer vision techniques provides the opportunity for the quantification of PTC attributes using conventional stains and whole-slide images. Methods: To explore the relationship between PTC characteristics and clinical outcome, n=280 periodic acid–Schiff-stained kidney biopsies (88 minimal change disease, 109 focal segmental glomerulosclerosis, 46 membranous nephropathy, and 37 IgA nephropathy) from the Nephrotic Syndrome Study Network digital pathology repository were computationally analyzed. A previously validated deep learning model was applied to segment cortical PTCs. Average PTC aspect ratio (PTC major to minor axis ratio), size (PTC pixels per PTC segmentation), and density (PTC pixels per unit cortical area) were computed for each biopsy. Cox proportional hazards models were used to assess associations between these PTC parameters and outcome (40% eGFR decline or kidney failure). Cortical PTC characteristics and interstitial fractional space PTC density were compared between areas of interstitial fibrosis and tubular atrophy (IFTA) and areas without IFTA. Results: When normalized PTC aspect ratio was below 0.6, a 0.1, increase in normalized PTC aspect ratio was significantly associated with disease progression, with a hazard ratio (95% confidence interval) of 1.28 (1.04 to 1.59) (P = 0.019), while PTC density and size were not significantly associated with outcome. Interstitial fractional space PTC density was lower in areas of IFTA compared with non-IFTA areas. Conclusions: Computational image analysis enables quantification of the status of the kidney microvasculature and the discovery of a previously unrecognized PTC biomarker (aspect ratio) of clinical outcome.