AI Article Synopsis

  • Researchers are tackling the challenges of finding early-stage markers and effectively treating melanoma by using deep learning to analyze standard pathology slides instead of costly transcriptome data.
  • They developed models that classify these slides into immune subgroups and achieved promising results, showing a strong ability to predict patient survival outcomes based on their immune classification.
  • This approach could enhance clinical workflows by identifying important biomarkers and helping clinicians understand tumor immune landscapes without the need for expensive genetic tests.

Article Abstract

Determining early-stage prognostic markers and stratifying patients for effective treatment are two key challenges for improving outcomes for melanoma patients. Previous studies have used tumour transcriptome data to stratify patients into immune subgroups, which were associated with differential melanoma specific survival and potential predictive biomarkers. However, acquiring transcriptome data is a time-consuming and costly process. Moreover, it is not routinely used in the current clinical workflow. Here, we attempt to overcome this by developing deep learning models to classify gigapixel haematoxylin and eosin (H&E) stained pathology slides, which are well established in clinical workflows, into these immune subgroups. We systematically assess six different multiple instance learning (MIL) frameworks, using five different image resolutions and three different feature extraction methods. We show that pathology-specific self-supervised models using 10x resolution patches generate superior representations for the classification of immune subtypes. In addition, in a primary melanoma dataset, we achieve a mean area under the receiver operating characteristic curve (AUC) of 0.80 for classifying histopathology images into 'high' or 'low immune' subgroups and a mean AUC of 0.82 in an independent TCGA melanoma dataset. Furthermore, we show that these models are able to stratify patients into 'high' and 'low immune' subgroups with significantly different melanoma specific survival outcomes (log rank test, P< 0.005). We anticipate that MIL methods will allow us to find new biomarkers of high importance, act as a tool for clinicians to infer the immune landscape of tumours and stratify patients, without needing to carry out additional expensive genetic tests.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.media.2024.103097DOI Listing

Publication Analysis

Top Keywords

stratify patients
12
multiple instance
8
instance learning
8
transcriptome data
8
immune subgroups
8
melanoma specific
8
specific survival
8
melanoma dataset
8
'high' 'low
8
'low immune'
8

Similar Publications

Electroconvulsive therapy modulates brain functional stability in patients with major depressive disorder.

J Affect Disord

January 2025

Department of Neurology, The First Affiliated Hospital of Anhui Medical University, 218 Jixi Road, Anhui Province, Hefei 230022, China; Department of Neurology, The Second Affiliated Hospital of Anhui Medical University, Hefei 230601, China; Department of Psychology and Sleep Medicine, The Second Affiliated Hospital of Anhui Medical University, Hefei 230601, China. Electronic address:

Background: Electroconvulsive therapy (ECT) is an effective treatment for patients with major depressive disorder (MDD), but the underlying neuromodulatory mechanisms remain largely unknown. Functional stability represents a newly developed method based on the dynamic functional connectivity framework. This study aimed to explore ECT-evoked changes in functional stability and their relationship with clinical outcomes.

View Article and Find Full Text PDF

Thyroid hormones in systemic lupus erythematosus: the catalyst for disease progression?

Am J Med Sci

January 2025

Henan Provincial People's Hospital, Henan Provincial Clinical Research Center for Kidney Disease, Henan Key Laboratory of Nephrology and Immunology, Zhengzhou 450003, Henan, China. Electronic address:

Objective: The study aimed to investigate the impact of varying thyroid function statuses on clinical and laboratory indicators in patients with systemic lupus erythematosus (SLE).

Methods: A retrospective analysis was conducted on 258 patients with SLE, who were stratified according to thyroid function, renal involvement, and disease activity. The predictive value of thyroid hormones was evaluated using a receiver operating characteristic (ROC) curve.

View Article and Find Full Text PDF

Dietary Inflammatory Score and Incident Heart Failure in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study.

J Card Fail

January 2025

Division of General Internal Medicine, Department of Medicine, Weill Cornell Medicine, New York, NY; Division of Cardiology, Department of Medicine, Weill Cornell Medicine, New York, NY. Electronic address:

Background: Inflammation plays a key role in the development of heart failure (HF), and diet is a known modifiable factor that modulates systemic inflammation. The dietary inflammatory score (DIS) is a tool to quantify the inflammatory components of diet. We sought to determine whether the DIS is associated with incident HF events.

View Article and Find Full Text PDF

The effect of Torque Teno Virus (TTV) Infection on Clinical Outcomes, Genomic Integrity, and Mortality in COPD Patients.

Mech Ageing Dev

January 2025

San Raffaele University; Department of Human Sciences and Promotion of the Quality of Life, Via di Val Cannuta 247, 00166 Rome, Italy; Clinical and Molecular Epidemiology, IRCCS San Raffaele Roma, Via di Val Cannuta 247, 00166 Rome, Italy. Electronic address:

Introduction: Torque Teno Virus (TTV), an "orphan" virus with unclear pathology, has been associated with various diseases and immune dysfunctions. This study investigates the link between TTV viremia and clinical markers in patients with severe to very severe COPD undergoing respiratory rehabilitation.

Methods: We analyzed 102 elderly COPD patients, stratified by TTV viremia levels (< or ≥ 4 log10 copies/mL).

View Article and Find Full Text PDF

Research Question: Does endometrial preparation using a natural cycle lead to higher live birth rates (LBR) in single euploid frozen embryo transfers (FET) compared with programmed cycles, for women who are normal weight, overweight and obese.

Design: Retrospective study of 845 single euploid FETs from 688 couples. Patients were stratified by body mass index (BMI) into normal weight, overweight and obesity class I/II categories.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!