A5058533373- Radiomics and Machine Learning in Medical Imaging
- Lung Cancer Diagnosis and Treatment
- Bayesian Modeling and Causal Inference
- AI in cancer detection
- Artificial Intelligence in Healthcare and Education
- Privacy-Preserving Technologies in Data
- Machine Learning in Healthcare
- Health Systems, Economic Evaluations, Quality of Life
- Artificial Intelligence in Healthcare
- Data Quality and Management
- Topic Modeling
- Advanced Radiotherapy Techniques
- Medical Imaging Techniques and Applications
- Cardiac Health and Mental Health
- Rheumatoid Arthritis Research and Therapies
- Biomedical Text Mining and Ontologies
- Chronic Disease Management Strategies
- Retinal and Optic Conditions
- Ocular Diseases and Behçet’s Syndrome
- Retinal Diseases and Treatments
- Colorectal Cancer Screening and Detection
- Dementia and Cognitive Impairment Research
- Cryptography and Data Security
- Healthcare Systems and Public Health
- Global Health Care Issues
Maastro Clinic
2019-2025
Maastricht University Medical Centre
2020-2025
Hasselt University
2024-2025
Maastricht University
2022-2024
University Medical Center
2023-2024
Centraal Bureau voor de Statistiek
2022
University of Sheffield
2017-2019
National University of Distance Education
2013-2018
Universidad Politécnica de Madrid
2017
National Institute of Astrophysics, Optics and Electronics
2017
Early diagnosis of lung cancer is a key intervention for the treatment in which computer-aided (CAD) can play crucial role. Most published CAD methods perform by classifying each nodule isolation. However, this does not reflect clinical practice, where clinicians diagnose patient based on set images nodules, instead looking at one time. Besides, low interpretability output provided these presents an important barrier their adoption.
Manual cohort building from radiology reports can be tedious. Natural Language Processing (NLP) used for automated building. In this study, we have developed and validated an NLP approach based on deep learning (DL) to select lung cancer a thoracic disease management group cohort. 4064 (CT PET/CT) of reported between 2014 2016 were used. These anonymised, cleaned, text normalized split into training, testing, validation set. External was performed the MIMIC-III clinical database. We three DL...
Establishing collaborations between cohort studies has been fundamental for progress in health research. However, such are hampered by heterogeneous data representations across cohorts and legal constraints to sharing. The first arises from a lack of consensus standards collection representation is usually tackled applying harmonization processes. second increasingly important due raised awareness privacy protection stricter regulations, as the GDPR. Federated learning emerged...
Abstract Background In Crohn’s disease (CD), artificial intelligence (AI) may improve treatment optimization and aid in clinical decision-making 1. The aim of the study is to compare results Bayesian Networks (BNs) both Expert Knowledge Model (EKM) Computer Algorithm Generated (CAGM) predicting corticosteroid-free remission at 52 weeks after introducing ustekinumab vedolizumab patients with CD. Methods Data were extracted from Dutch Initiative on Crohn Colitis (ICC) registry. Observations...
Lung cancer (LC) is the top cause of deaths globally, prompting many countries to adopt LC screening programs. While typically relies on age and smoking intensity, more efficient risk models exist. We devised a Bayesian network (BN) for detection, testing its resilience with varying degrees missing data comparing it prior machine learning (ML) model. analyzed from 9940 patients referred assessment in Southern Denmark 2009 2018. Variables included age, sex, smoking, lab results. Our...
Aggregation of cohort data increases precision for studying neurodegenerative disease pathways, but efforts to combine and expertise are often hampered by infrastructural, ethical legal considerations. We aimed unite various studies in the Netherlands enhance research infrastructure facilitate on dementia etiology its public health implications. The Consortium Dementia Cohorts (NCDC) includes participants with initially no established cognitive impairment from 9 Dutch cohorts: Amsterdam...
Background Accurate prediction of pathologic complete response (pCR) following neoadjuvant immunotherapy combined with chemotherapy (nICT) is crucial for tailoring patient care in esophageal squamous cell carcinoma (ESCC). This study aimed to develop and validate a deep learning model using novel voxel-level radiomics approach predict pCR based on preoperative CT images. Methods In this multicenter, retrospective study, 741 patients ESCC who underwent nICT followed by radical esophagectomy...
Rising incidence and mortality of cancer have led to an incremental amount research in the field. To learn from preexisting data, it has become important capture maximum information related disease type, stage, treatment, outcomes. Medical imaging reports are rich this kind but only present as free text. The extraction such unstructured text is labor-intensive. use Natural Language Processing (NLP) tools extract radiology can make less time-consuming well more effective. In study, we...
Objective Cardiovascular diseases (CVD) are one of the most prevalent in India amounting for nearly 30% total deaths. A dearth research on CVD risk scores Indian population, limited performance conventional and inability to reproduce initial accuracies randomised clinical trials has led this study large-scale patient data. The objective is develop an Artificial Intelligence-based Risk Score (AICVD) predict event (eg, acute myocardial infarction/acute coronary syndrome) next 10 years compare...
Background Natural language processing (NLP) is thought to be a promising solution extract and store concepts from free text in structured manner for data mining purposes. This also true radiology reports, which still consist mostly of text. Accurate complete reports are very important clinical decision support, instance, oncological staging. As such, NLP can tool structure the content report, thereby increasing report’s value. Objective study describes implementation validation an N-stage...
Objective: The software application FOX (‘Fitting to Outcome eXpert’) is an intelligent agent assist in the programing of cochlear implant (CI) processors. current version utilizes a mixture deterministic and probabilistic logic which able improve over time through learning effect. This study aimed at assessing whether this capacity yields measurable improvements speech understanding.Methods: A retrospective was performed on 25 consecutive CI recipients with median use experience 10 years...
Abstract Background Cancer prognosis before and after treatment is key for patient management decision making. Handcrafted imaging biomarkers—radiomics—have shown potential in predicting prognosis. Purpose However, given the recent progress deep learning, it timely relevant to pose question: could learning based 3D features be used as biomarkers outperform radiomics? Methods Effectiveness, reproducibility test/retest, across modalities, correlation of with clinical such tumor volume TNM...
Privacy-preserving machine learning enables the training of models on decentralized datasets without need to reveal information, both horizontally and vertically partitioned data. However, it requires specialized techniques algorithms perform necessary computations. The privacy preserving scalar product protocol, which dot vectors revealing them, is one popular example for its versatility. For can be used analyses that require counting number samples fulfill certain criteria defined across...
Abstract Background As a means to extract biomarkers from medical imaging, radiomics has attracted increased attention researchers. However, reproducibility and performance of in low‐dose CT scans are still poor, mostly due noise. Deep learning generative models can be used denoise these images turn improve radiomics’ performance. most trained on paired data, which difficult or impossible collect. Purpose In this article, we investigate the possibility denoising CTs using cycle adversarial...
Radiomics is an active area of research focusing on high throughput feature extraction from medical images with a wide array applications in clinical practice, such as decision support oncology. However, noise low dose computed tomography (CT) scans can impair the accurate radiomic features. In this article, we investigate possibility using deep learning generative models to improve performance radiomics CTs.We used two datasets CT - NSCLC Radiogenomics and LIDC-IDRI test for tasks...
PURPOSE Randomized controlled trials are considered the golden standard for estimating treatment effect but costly to perform and not always possible. Observational data, although readily available, is sensitive biases such as confounding by indication. Structure learning algorithms Bayesian Networks (BNs) can be used discover underlying model from data. This enables identification of confounders through graph analysis, might contain noncausal edges. We propose using a blacklist aid...
Artificial intelligence applications in radiation oncology have been the focus of study last decade. The introduction automated and intelligent solutions for routine clinical tasks, such as treatment planning quality assurance, has potential to increase safety efficiency radiotherapy. In this work, we present a multi-institutional across three different institutions internationally on Bayesian network (BN)-based initial plan review assistive tool that alerts radiotherapy professionals...
Introduction Urinary incontinence (UI) is a common side effect of prostate cancer treatment, but in clinical practice, it difficult to predict. Machine learning (ML) models have shown promising results predicting outcomes, yet the lack transparency complex known as “black-box” has made clinicians wary relying on them sensitive decisions. Therefore, finding balance between accuracy and explainability crucial for implementation ML models. The aim this study was employ three different...