MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification

Scientific Data, Journal Year: 2023, Volume and Issue: 10(1)

Published: Jan. 19, 2023

Abstract We introduce MedMNIST v2 , a large-scale MNIST-like dataset collection of standardized biomedical images, including 12 datasets for 2D and 6 3D. All images are pre-processed into small size 28 × (2D) or (3D) with the corresponding classification labels so that no background knowledge is required users. Covering primary data modalities in designed to perform on lightweight 3D various scales (from 100 100,000) diverse tasks (binary/multi-class, ordinal regression, multi-label). The resulting dataset, consisting 708,069 9,998 total, could support numerous research/educational purposes image analysis, computer vision, machine learning. benchmark several baseline methods v2, 2D/3D neural networks open-source/commercial AutoML tools. code publicly available at https://medmnist.com/ .

Language: Английский

---

The Application of Deep Learning in Cancer Prognosis Prediction

Cancers, Journal Year: 2020, Volume and Issue: 12(3), P. 603 - 603

Published: March 5, 2020

Deep learning has been applied to many areas in health care, including imaging diagnosis, digital pathology, prediction of hospital admission, drug design, classification cancer and stromal cells, doctor assistance, etc. Cancer prognosis is estimate the fate cancer, probabilities recurrence progression, provide survival estimation patients. The accuracy will greatly benefit clinical management improvement biomedical translational research application advanced statistical analysis machine methods are driving forces improve prediction. Recent years, there a significant increase computational power rapid advancement technology artificial intelligence, particularly deep learning. In addition, cost reduction large scale next-generation sequencing, availability such data through open source databases (e.g., TCGA GEO databases) offer us opportunities possibly build more powerful accurate models predict accurately. this review, we reviewed most recent published works that used for suggested be generic model, requires less engineering, achieves when working with amounts data. shown equivalent or better than current approaches, as Cox-PH. With burst multi-omics data, genomics transcriptomics information studies, believe would potentially prognosis.

Language: Английский

---

Emerging role of deep learning‐based artificial intelligence in tumor pathology

Cancer Communications, Journal Year: 2020, Volume and Issue: 40(4), P. 154 - 166

Published: April 1, 2020

Abstract The development of digital pathology and progression state‐of‐the‐art algorithms for computer vision have led to increasing interest in the use artificial intelligence (AI), especially deep learning (DL)‐based AI, tumor pathology. DL‐based been developed conduct all kinds work involved pathology, including diagnosis, subtyping, grading, staging, prognostic prediction, as well identification pathological features, biomarkers genetic changes. applications AI not only contribute improve diagnostic accuracy objectivity but also reduce workload pathologists subsequently enable them spend additional time on high‐level decision‐making tasks. In addition, is useful meet requirements precision oncology. However, there are still some challenges relating implementation issues algorithm validation interpretability, computing systems, unbelieving attitude pathologists, clinicians patients, regulators reimbursements. Herein, we present an overview how AI‐based approaches could be integrated into workflow discuss perspectives

Language: Английский

---

Deep learning model for the prediction of microsatellite instability in colorectal cancer: a diagnostic study

The Lancet Oncology, Journal Year: 2020, Volume and Issue: 22(1), P. 132 - 141

Published: Dec. 30, 2020

Language: Английский

---

Pan-cancer integrative histology-genomic analysis via multimodal deep learning

Cancer Cell, Journal Year: 2022, Volume and Issue: 40(8), P. 865 - 878.e6

Published: Aug. 1, 2022

The rapidly emerging field of computational pathology has demonstrated promise in developing objective prognostic models from histology images. However, most are either based on or genomics alone and do not address how these data sources can be integrated to develop joint image-omic models. Additionally, identifying explainable morphological molecular descriptors that govern such prognosis is interest. We use multimodal deep learning jointly examine whole-slide images profile 14 cancer types. Our weakly supervised, deep-learning algorithm able fuse heterogeneous modalities predict outcomes discover features correlate with poor favorable outcomes. present all analyses for correlates patient across the types at both a disease level an interactive open-access database allow further exploration, biomarker discovery, feature assessment.

Language: Английский

---

Clinical-Grade Detection of Microsatellite Instability in Colorectal Tumors by Deep Learning

Gastroenterology, Journal Year: 2020, Volume and Issue: 159(4), P. 1406 - 1416.e11

Published: June 17, 2020

Microsatellite instability (MSI) and mismatch-repair deficiency (dMMR) in colorectal tumors are used to select treatment for patients. Deep learning can detect MSI dMMR tumor samples on routine histology slides faster less expensively than molecular assays. However, clinical application of this technology requires high performance multisite validation, which have not yet been performed.

Language: Английский

---

Transformer-based unsupervised contrastive learning for histopathological image classification

Medical Image Analysis, Journal Year: 2022, Volume and Issue: 81, P. 102559 - 102559

Published: July 30, 2022

Language: Английский

---

Designing deep learning studies in cancer diagnostics

Nature reviews. Cancer, Journal Year: 2021, Volume and Issue: 21(3), P. 199 - 211

Published: Jan. 29, 2021

Language: Английский

---

Artificial intelligence in histopathology: enhancing cancer research and clinical oncology

Nature Cancer, Journal Year: 2022, Volume and Issue: 3(9), P. 1026 - 1038

Published: Sept. 22, 2022

Language: Английский

---

Predicting Survival After Hepatocellular Carcinoma Resection Using Deep Learning on Histological Slides

Hepatology, Journal Year: 2020, Volume and Issue: 72(6), P. 2000 - 2013

Published: Feb. 28, 2020

Standardized and robust risk-stratification systems for patients with hepatocellular carcinoma (HCC) are required to improve therapeutic strategies investigate the benefits of adjuvant systemic therapies after curative resection/ablation.In this study, we used two deep-learning algorithms based on whole-slide digitized histological slides (whole-slide imaging; WSI) build models predicting survival HCC treated by surgical resection. Two independent series were investigated: a discovery set (Henri Mondor Hospital, n = 194) develop our an validation (The Cancer Genome Atlas [TCGA], 328). WSIs first divided into small squares ("tiles"), features extracted pretrained convolutional neural network (preprocessing step). The deep-learning-based algorithm ("SCHMOWDER") uses attention mechanism tumoral areas annotated pathologist whereas second ("CHOWDER") does not require human expertise. In set, c-indices prediction SCHMOWDER CHOWDER reached 0.78 0.75, respectively. Both outperformed composite score incorporating all baseline variables associated survival. Prognostic value was further validated in TCGA data and, as observed series, both had higher discriminatory power than combining Pathological review showed that most predictive poor characterized vascular spaces, macrotrabecular architectural pattern, lack immune infiltration.This study shows artificial intelligence can help refine prognosis. It highlights importance pathologist/machine interactions construction benefit from expert knowledge allow biological understanding their output.

Language: Английский

---