A5062929705- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Lung Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- Glioma Diagnosis and Treatment
- Advanced X-ray and CT Imaging
- Cancer-related molecular mechanisms research
- Molecular Biology Techniques and Applications
- Machine Learning in Healthcare
- Breast Cancer Treatment Studies
- Gene expression and cancer classification
- Cancer, Hypoxia, and Metabolism
- Myeloproliferative Neoplasms: Diagnosis and Treatment
- Chronic Myeloid Leukemia Treatments
- Hepatocellular Carcinoma Treatment and Prognosis
- Prostate Cancer Treatment and Research
- Brain Tumor Detection and Classification
- Colorectal Cancer Screening and Detection
- Colorectal Cancer Surgical Treatments
- Gastric Cancer Management and Outcomes
- Machine Learning in Materials Science
- Sarcoma Diagnosis and Treatment
- Artificial Intelligence in Healthcare and Education
- Radiopharmaceutical Chemistry and Applications
- Multiple Myeloma Research and Treatments
Cedars-Sinai Medical Center
2023-2024
University Hospital Carl Gustav Carus
2020-2024
Helmholtz-Zentrum Dresden-Rossendorf
2020-2024
TU Dresden
2020-2024
National Center for Tumor Diseases
2020-2024
German Cancer Research Center
2020-2024
Cardiff University
2024
University of Pennsylvania
2024
University of California, San Francisco
2020-2023
Université de Bretagne Occidentale
2015-2020
The image biomarker standardisation initiative (IBSI) is an independent international collaboration which works towards standardising the extraction of biomarkers from acquired imaging for purpose high-throughput quantitative analysis (radiomics). Lack reproducibility and validation studies considered to be a major challenge field. Part this lies in scantiness consensus-based guidelines definitions process translating into biomarkers. IBSI therefore seeks provide nomenclature definitions,...
Machine learning (ML) algorithms for selecting and combining radiomic features into multiparametric prediction models have become popular; however, it has been shown that large variations in performance can be obtained by relying on different approaches. The purpose of this study was to evaluate the potential benefit an improved consensus final prediction, as other fields. Methods: evaluation carried out context use radiomics from 18F-FDG PET/CT images predicting outcome stage II-III...
Tumor hypoxia increases resistance to radiotherapy and systemic therapy. Our aim was develop validate a disease-agnostic disease-specific CT (+FDG-PET) based radiomics classification signature.A total of 808 patients with imaging data were included: N = 100 training/N 183 external validation cases for signature, 76 39 the H&N signature 62 36 Lung signature. The primary gross tumor volumes (GTV) manually defined by experts on CT. In order dichotomize between hypoxic/well-oxygenated tumors...
Machine learning techniques are becoming increasingly popular in radiomics studies. They can handle high dimensional sets of features with higher robustness than usual statistical analyses, by capturing complex interactions between themselves and feature combinations clinical endpoints under investigation order to build efficient prognostic/predictive models. However, there is no "one fits all" solution deciding which algorithm the most accurate for a given application not always...
Image-derived features (“radiomics”) are increasingly being considered for patient management in (neuro)oncology and radiotherapy. In Glioblastoma multiforme (GBM), simple often used by clinicians clinical practice, such as the size of tumor or relative sizes necrosis active tumor. First order statistics provide a limited characterization power because they do not incorporate spatial information thus cannot differentiate patterns. this work, we present methodological framework building...
Heterogeneity image-derived features of Glioblastoma multiforme (GBM) tumors from multimodal MRI sequences may provide higher prognostic value than standard parameters used in routine clinical practice. We previously developed a framework for automatic extraction and combination (also called "Radiomics") through support vector machines (SVM) predictive model building. The results we obtained cohort 40 GBM suggested these could be to identify patients with poorer outcome. However, is delicate...
Background: The high dimensionality of radiomic feature sets, the variability in types and potentially computational requirements all underscore need for an effective method to identify smallest set predictive features a given clinical problem. Purpose: Develop methodology tools explain features. Materials Methods: 89,714 were extracted from five cancer datasets: low-grade glioma, meningioma, non-small cell lung (NSCLC), two renal carcinoma cohorts (n=2104). Features categorized by...
Abstract Deep learning transformer models have exhibited exceptional performance in various clinical tasks, including cancer outcome prediction, when applied to electronic health records (EHR). Inspired by the success of bidirectional encoder representations from transformers (BERT) natural language processing, we present OncoBERT, a deep transfer framework tailored for prediction using unstructured notes diverse sites Glioma, Prostate, and Breast. OncoBERT adapts BERT EHR processing employs...
In August 2022, the Cancer Informatics for Centers brought together cancer informatics leaders its biannual symposium, Precision Medicine Applications in Radiation Oncology, co-chaired by Quynh-Thu Le, MD (Stanford University), and Walter J. Curran, (GenesisCare). Over course of 3 days, presenters discussed a range topics relevant to radiation oncology community more broadly, including biomarker development, decision support algorithms, novel imaging tools, theranostics, artificial...
Abstract Objectives To apply CT-based foundational artificial intelligence (AI) and radiomics models for predicting overall survival (OS) patients with locally advanced non-small cell lung cancer (NSCLC). Methods Data 449 retrospectively treated on the NRG Oncology/Radiation Therapy Oncology Group (RTOG) 0617 clinical trial were analyzed. Foundational AI, radiomics, features evaluated using univariate cox regression correlational analyses to determine independent predictors of survival....
Recurrence occurs in more than 50% of prostate cancer. To be effective, treatments require precise localization tumor cells. [F]fluoromethylcholine ([18F]FCH) PET/computed tomography (CT) is currently used to restage disease cases biochemical relapse. for therapy response as has been suggested, repeatability limits PET derived indices need established.The aim our study was prospectively assess the qualitative and quantitative reproducibility [18F]FCH PET/CT cancer.Patients with...
Abstract OBJECTIVE. A clinical challenge of ductal carcinoma in situ (DCIS) is to accurately predict individual risk for recurrence. Prior studies have highlighted the importance stromal collagen surrounding breast ducts cancer aggression. We studied structure DCIS a matched case-control cohort women with using second-harmonic generation (SHG) and hematoxylin eosin (H&E) images determine whether could provide additional information. METHODS. 122 patients (61 cases recurrence 61 controls)...
Abstract OBJECTIVE It remains a challenge to predict individual risk for recurrence after primary treatment of ductal carcinoma in situ (DCIS). While DCIS is contained within the duct, prior studies have pointed importance stromal collagen cancer progression. Using hematoxylin and eosin (H&E) stain-based images, we studied structure surrounding case-control cohort women determine whether could provide additional information. METHODS We present quantitative histology image analysis...