Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
The cover shows image patches cropped from slides that were used as training input in a deep learning system for gynecologic cytopathology. For more information, see the paper by Ke et al, this issue (p 513).
Data processing and learning has become a spearhead for the advancement of medicine. Computational pathology is burgeoning subspecialty that promises a better-integrated solution to whole-slide images, multi-omics data and clinical informatics as innovative approach for patient care. This review describes clinical perspectives and discusses the statistical methods, clinical applications, potential obstacles, and future directions of computational pathology.
Tandem mass spectrometry can reveal metabolite positional labeling and improve the performance of metabolic flux analysis as long as daughter ions are carefully inspected. When calculating the fluxes, the tandem mass isotopomer distributions as well as the mass isotopomer distributions of parent and daughter ions should all be used to constrain the fluxes in order to achieve the best performance.
Most current biomedical datasets are rectangular in shape and have few missing data, but the sample sizes are very large. Rigorous analyses of these huge datasets demand considerably more efficient and more accurate machine-learning algorithms to classify outcomes. This paper aims to determine the performance and efficiency of classifying multi-category outcomes of such rectangular data.
The authors present an assessment of the utility of “third-generation” nanopore sequencing for the confirmation and characterization of mobile element insertions. and discuss how implementation of long-read nanopore sequencing can offer benefits over existing molecular approaches.
In this paper, the authors describe the development and validation of a novel image signature-based radiomics model. A total of 655 glioma patients were enrolled to build this model which is shown to be an effective tool to achieve multilayer preoperative diagnosis and prognostic stratification of gliomas.
The correlation between SMAD4 mutations and clinico-molecular features in a Chinese non-small cell lung cancer (NSCLC) cohort was studied using next-generation sequencing. Integrated bioinformatics analyses based on public databases were conducted to further investigate the prognostic value of SMAD4. Results showed that SMAD4 alteration was associated with poor survival and resistance to platinum-based chemotherapy, suggesting that SMAD4 alteration might be a predictive marker in NSCLC.
In this study, effect of Osr1 deficiency in promoting non-alcoholic fatty liver disease progression was examined. The authors demonstrate that Osr1 regulates hepatic inflammation and cell survival through multiple signaling pathways and DNA methylation modification.
Computational modeling has emerged as a promising and cost-effective alternative method for screening potentially endocrine active compounds. This study applies classic machine learning algorithms and deep learning approaches to a panel of over 7500 compounds tested against 18 Toxicity Forecaster assays related to nuclear estrogen receptor (ERα and ERβ) activity.
The authors show that genomes of female infertility patients harbor mutations within miRNA genes and 3’ UTR miRNA binding sites that are likely to affect maternal transcript clearance. Such disruption may not be compatible with zygotic genome activation and could cause a failure in embryonic development, implantation or cause early miscarriage, resulting in an infertility diagnosis.
This manuscript describes a methodology to quantify the abnormalities in digital cytology images. This automatic AI-system incorporates deep learning structures, mathematical algorithms, and image processing methods to locate and segment abnormal and suspicious cells. Characterized as more informative, objective, and reproducible, it has the potential to assist clinical practice.
The authors developed a deep-learning-based ductal carcinoma in situ (DCIS) grading system that achieved performance similar to expert observers. To the best of our knowledge, this is the first automated system for the grading of DCIS that could assist pathologists by providing robust and reproducible second opinions on DCIS grade.