A5049138051- Radiomics and Machine Learning in Medical Imaging
- Lung Cancer Diagnosis and Treatment
- Advanced X-ray and CT Imaging
- Medical Imaging Techniques and Applications
- AI in cancer detection
- Cancer Genomics and Diagnostics
- Lung Cancer Treatments and Mutations
- COVID-19 diagnosis using AI
- Gastric Cancer Management and Outcomes
- Colorectal Cancer Treatments and Studies
- Ferroptosis and cancer prognosis
- Sarcoma Diagnosis and Treatment
- Cardiac Imaging and Diagnostics
- Molecular Biology Techniques and Applications
- Gene expression and cancer classification
- MRI in cancer diagnosis
- Artificial Intelligence in Healthcare and Education
- Inflammatory Biomarkers in Disease Prognosis
- Mindfulness and Compassion Interventions
- Problem and Project Based Learning
- Generative Adversarial Networks and Image Synthesis
- Histiocytic Disorders and Treatments
- Nanoparticle-Based Drug Delivery
- Adipokines, Inflammation, and Metabolic Diseases
- Bone fractures and treatments
Novartis Institutes for BioMedical Research
2022
Novartis (Switzerland)
2022
Harvard University
2013-2021
Mass General Brigham
2021
Brigham and Women's Hospital
2013-2020
Dana-Farber Cancer Institute
2013-2020
Dana-Farber Brigham Cancer Center
2013-2019
Maharaja Sayajirao University of Baroda
2018
Maastricht University
2013-2017
Maastro Clinic
2013-2017
Radiomics aims to quantify phenotypic characteristics on medical imaging through the use of automated algorithms. Radiomic artificial intelligence (AI) technology, either based engineered hard-coded algorithms or deep learning methods, can be used develop noninvasive imaging-based biomarkers. However, lack standardized algorithm definitions and image processing severely hampers reproducibility comparability results. To address this issue, we developed PyRadiomics, a flexible open-source...
Human cancers exhibit strong phenotypic differences that can be visualized noninvasively by medical imaging. Radiomics refers to the comprehensive quantification of tumour phenotypes applying a large number quantitative image features. Here we present radiomic analysis 440 features quantifying intensity, shape and texture, which are extracted from computed tomography data 1,019 patients with lung or head-and-neck cancer. We find have prognostic power in independent sets cancer patients, many...
Radiomics extracts and mines large number of medical imaging features quantifying tumor phenotypic characteristics. Highly accurate reliable machine-learning approaches can drive the success radiomic applications in clinical care. In this study, fourteen feature selection methods twelve classification were examined terms their performance stability for predicting overall survival. A total 440 extracted from pre-treatment computed tomography (CT) images 464 lung cancer patients. To ensure...
Due to advances in the acquisition and analysis of medical imaging, it is currently possible quantify tumor phenotype. The emerging field Radiomics addresses this issue by converting images into minable data extracting a large number quantitative imaging features. One main challenges segmentation. Where manual delineation time consuming prone inter-observer variability, has been shown that semi-automated approaches are fast reduce variability. In study, semiautomatic region growing...
Non-small-cell lung cancer (NSCLC) patients often demonstrate varying clinical courses and outcomes, even within the same tumor stage. This study explores deep learning applications in medical imaging allowing for automated quantification of radiographic characteristics potentially improving patient stratification.We performed an integrative analysis on 7 independent datasets across 5 institutions totaling 1,194 NSCLC (age median = 68.3 years [range 32.5-93.3], survival 1.7 0.0-11.7]). Using...
Abstract Purpose: Tumors are continuously evolving biological systems, and medical imaging is uniquely positioned to monitor changes throughout treatment. Although qualitatively tracking lesions over space time may be trivial, the development of clinically relevant, automated radiomics methods that incorporate serial data far more challenging. In this study, we evaluated deep learning networks for predicting clinical outcomes through analyzing series CT images patients with locally advanced...
IntroductionImmunotherapy is regarded as one of the major breakthroughs in cancer treatment. Despite its success, only a subset patients responds—urging quest for predictive biomarkers. We hypothesize that artificial intelligence (AI) algorithms can automatically quantify radiographic characteristics are related to and may therefore act noninvasive radiomic biomarkers immunotherapy response.Patients methodsIn this study, we analyzed 1055 primary metastatic lesions from 203 with advanced...
Abstract Radiomics provides a comprehensive quantification of tumor phenotypes by extracting and mining large number quantitative image features. To reduce the redundancy compare prognostic characteristics radiomic features across cancer types, we investigated cancer-specific feature clusters in four independent Lung Head & Neck (H&N) cohorts (in total 878 patients). Radiomic were extracted from pre-treatment computed tomography (CT) images. Consensus clustering resulted eleven...
Purpose. Besides basic measurements as maximum standardized uptake value (SUV)max or SUVmean derived from 18F-FDG positron emission tomography (PET) scans, more advanced quantitative imaging features (i.e. “Radiomics” features) are increasingly investigated for treatment monitoring, outcome prediction, potential biomarkers. With these prospected applications of Radiomics features, it is a requisite that they provide robust and reliable measurements. The aim our study was therefore to perform...
Tumors are characterized by somatic mutations that drive biological processes ultimately reflected in tumor phenotype. With regard to radiographic phenotypes, generally unconnected through present understanding the presence of specific mutations, artificial intelligence methods can automatically quantify phenotypic characters using predefined, engineered algorithms or automatic deep-learning methods, a process also known as radiomics. Here we demonstrate how imaging phenotypes be connected...
"Radiomics" extracts and mines a large number of medical imaging features in non-invasive cost-effective way. The underlying assumption radiomics is that these quantify phenotypic characteristics an entire tumor. In order to enhance applicability clinical oncology, highly accurate reliable machine-learning approaches are required. this radiomic study, 13 feature selection methods 11 classification were evaluated terms their performance stability for predicting overall survival head neck...
Radiomics can quantify tumor phenotypic characteristics non-invasively by applying feature algorithms to medical imaging data. In this study of lung cancer patients, we investigated the association between radiomic features and histologic subtypes (adenocarcinoma squamous cell carcinoma). Furthermore, in order predict subtypes, employed machine-learning methods independently evaluated their prediction performance.Two independent cohorts with a combined size 350 patients were included our...
Medical imaging can visualize characteristics of human cancer noninvasively. Radiomics is an emerging field that translates these medical images into quantitative data to enable phenotypic profiling tumors. While radiomics has been associated with several clinical endpoints, the complex relationships radiomics, factors, and tumor biology are largely unknown. To this end, we analyzed two independent cohorts respectively 262 North American 89 European patients lung cancer, consistently...
Abstract Multiparametric Magnetic Resonance Imaging (MRI) can provide detailed information of the physical characteristics rectum tumours. Several investigations suggest that volumetric analyses on anatomical and functional MRI contain clinically valuable information. However, manual delineation tumours is a time consuming procedure, as it requires high level expertise. Here, we evaluate deep learning methods for automatic localization segmentation rectal cancers multiparametric MR imaging....
Accurate volumetric assessment in non-small cell lung cancer (NSCLC) is critical for adequately informing treatments. In this study we assessed the clinical relevance of a semiautomatic computed tomography (CT)-based segmentation method using competitive region-growing based algorithm, implemented free and public available 3D-Slicer software platform. We compared segmented volumes by three independent observers, who primary tumour 20 NSCLC patients twice, to manual slice-by-slice...
Abstract Coronary artery calcium is an accurate predictor of cardiovascular events. While it visible on all computed tomography (CT) scans the chest, this information not routinely quantified as requires expertise, time, and specialized equipment. Here, we show a robust time-efficient deep learning system to automatically quantify coronary routine cardiac-gated non-gated CT. As evaluate in 20,084 individuals from distinct asymptomatic (Framingham Heart Study, NLST) stable acute chest pain...
PET-based radiomics have been used to noninvasively quantify the metabolic tumor phenotypes; however, little is known about relationship between these phenotypes and underlying somatic mutations. This study assessed association predictive power of <sup>18</sup>F-FDG PET–based radiomic features for mutations in non–small cell lung cancer patients. <b>Methods:</b> Three hundred forty-eight patients underwent diagnostic PET scans were tested genetic Thirteen percent (44/348) 28% (96/348) found...
Objectives The clinical management of meningioma is guided by tumor grade and biological behavior. Currently, the assessment follows surgical resection histopathologic review. Reliable techniques for pre-operative determination may enhance decision-making. Methods A total 175 patients (103 low-grade 72 high-grade) with contrast-enhanced T1-MRI were included. Fifteen radiomic (quantitative) 10 semantic (qualitative) features applied to quantify imaging phenotype. Area under curve (AUC) odd...
Methods from the field of machine (deep) learning have been successful in tackling a number tasks medical imaging, image reconstruction or processing to predictive modeling, clinical planning and decision-aid systems. The ever growing availability data improving ability algorithms learn them has led rise methods based on neural networks address most these with higher efficiency often superior performance than previous, "shallow" methods. present editorial aims at contextualizing within this...
Tumor phenotypes captured in computed tomography (CT) images can be described qualitatively and quantitatively using radiologist-defined "semantic" computer-derived "radiomic" features, respectively. While both types of features have shown to promising predictors prognosis, the association between these groups remains unclear. We investigated associations semantic radiomic CT 258 non-small cell lung adenocarcinomas. The tumor imaging were 9 qualitative that scored by radiologists, 57...
Radiomics aims to quantitatively capture the complex tumor phenotype contained in medical images associate them with clinical outcomes. This study investigates impact of different types computed tomography (CT) on prognostic performance radiomic features for disease recurrence early stage non-small cell lung cancer (NSCLC) patients treated stereotactic body radiation therapy (SBRT). 112 NSCLC SBRT that had static free breathing (FB) and average intensity projection (AIP) were analyzed....
Cell lines and patient-derived xenografts are essential to cancer research; however, the results derived from such models often lack clinical translatability, as they do not fully recapitulate complex biology. Identifying preclinical that sufficiently resemble biological characteristics of tumors across different cancers is critically important. Here, we developed MOBER, Multi-Origin Batch Effect Remover method, simultaneously extract biologically meaningful embeddings while removing...