A5101901305- Radiomics and Machine Learning in Medical Imaging
- Advanced Radiotherapy Techniques
- Prostate Cancer Diagnosis and Treatment
- Lung Cancer Diagnosis and Treatment
- Renal cell carcinoma treatment
- Cancer Genomics and Diagnostics
- Economic and Financial Impacts of Cancer
- Artificial Intelligence in Healthcare and Education
- Prostate Cancer Treatment and Research
- Colorectal Cancer Screening and Detection
- Global Cancer Incidence and Screening
- Palliative Care and End-of-Life Issues
- Clinical Reasoning and Diagnostic Skills
- Lung Cancer Treatments and Mutations
- Bladder and Urothelial Cancer Treatments
- Glioma Diagnosis and Treatment
- Cervical Cancer and HPV Research
- Cancer Immunotherapy and Biomarkers
- Pain Management and Opioid Use
- Radiology practices and education
- Opioid Use Disorder Treatment
- Lung Cancer Research Studies
- Cardiac, Anesthesia and Surgical Outcomes
- Brain Metastases and Treatment
- Topic Modeling
The University of Texas Southwestern Medical Center
2023-2025
Kaiser Permanente
2024-2025
Southwestern Medical Center
2024-2025
Gangnam Severance Hospital
2023
Yonsei University
2023
Yale University
2013-2023
Princeton University
2015
University of California, San Francisco
2013
Bristol-Myers Squibb (United States)
2009
Chinese PLA General Hospital
1992
Importance Large language models (LLMs) have shown promise in their performance on both multiple-choice and open-ended medical reasoning examinations, but it remains unknown whether the use of such tools improves physician diagnostic reasoning. Objective To assess effect an LLM physicians’ compared with conventional resources. Design, Setting, Participants A single-blind randomized clinical trial was conducted from November 29 to December 29, 2023. Using remote video conferencing in-person...
With recent approval of standalone HPV testing and increasing uptake vaccination, some have postulated that we are moving toward a "post-Pap" era cervical cancer prevention. However, the total number cases been prevented by Pap smear screening as well its impact on racial disparities unknown.
<h3>Importance</h3> Cancer registries are important real-world data sources consisting of abstraction from the medical record; however, patients with unknown or missing underrepresented in studies that use such sources. <h3>Objective</h3> To assess prevalence and its association overall survival among cancer. <h3>Design, Setting, Participants</h3> In this retrospective cohort study, all variables within National Database were reviewed for values 3 most common cancers US who received...
Background Clinical trial matching, essential for advancing medical research, involves detailed screening of potential participants to ensure alignment with specific requirements. Research staff face challenges due the high volume eligible patients and complexity varying eligibility criteria. The traditional manual process, both time-consuming error-prone, often leads missed opportunities. Recently, large language models (LLMs), specifically generative pre-trained transformers (GPTs), have...
ABSTRACT Importance Diagnostic errors are common and cause significant morbidity. Large language models (LLMs) have shown promise in their performance on both multiple-choice open-ended medical reasoning examinations, but it remains unknown whether the use of such tools improves diagnostic reasoning. Objective To assess impact GPT-4 LLM physicians’ compared to conventional resources. Design Multi-center, randomized clinical vignette study. Setting The study was conducted using remote video...
BACKGROUND Ideally, screening detects cancer at a more curable stage and, as result, decreases the rate of subsequent diagnosis late stage. Although it is suggested that some tests have led to substantial increases in early‐stage incidence with only marginal reductions late‐stage (eg mammography), association between temporal trends colorectal and its cumulative impact on unknown. METHODS Colorectal data spanning over 3 decades (1976‐2009) were collected from Surveillance, Epidemiology, End...
<h3>Importance</h3> Prescription opioids are frequently prescribed to treat cancer-related pain. However, limited information exists regarding rates of prescription opioid use and misuse in populations with cancer. <h3>Objectives</h3> To estimate the prevalence likelihood adult cancer survivors compared respondents without identify characteristics associated survivors. <h3>Design, Setting, Participants</h3> This cross-sectional study is a retrospective, population-based using data from 169...
Deep learning (DL) models have rapidly become a popular and cost-effective tool for image classification within oncology. A major limitation of DL is their vulnerability to adversarial images, manipulated input images designed cause misclassifications by models. The purpose the study investigate robustness trained on diagnostic using explore utility an iterative training approach improve against images.We examined impact accuracies classify cancerous lesions across three common oncologic...
Abstract Background In the wake of US opioid epidemic, there have been efforts to curb prescribing. However, it is unknown whether these affected prescribing among oncologists, whose patients often require opioids for symptom management. We investigated temporal patterns in Medicare beneficiaries oncologists and nononcologists. Methods queried Centers Medicaid Services Part D prescriber dataset all physicians between January 1, 2013, December 31, 2017. used population-averaged multivariable...
Patients with cancer may be at risk of high opioid use due to physical and psychosocial factors, although little data exist inform providers policymakers. Our aim is examine overdoses from opioids leading emergency department (ED) visits among patients in the United States.
Large language model (LLM) artificial intelligence (AI) systems have shown promise in diagnostic reasoning, but their utility management reasoning with no clear right answers is unknown. To determine whether LLM assistance improves physician performance on open-ended tasks compared to conventional resources. Prospective, randomized controlled trial conducted from 30 November 2023 21 April 2024. Multi-institutional study Stanford University, Beth Israel Deaconess Medical Center, and the...
Abstract Introduction: Auto-segmentation of tumor volumes and organs at risk (OARs) is a critical step in cancer radiotherapy treatment planning, where rapid, precise adjustments to plans are required match the patient anatomy. Although auto-segmentation has been clinically accepted for most OARs, volumes, particularly clinical target (CTVs), remains challenge. This difficulty arises because images alone often insufficient capture necessary information accurate delineation microscopic...
758 Background: Antibody drug conjugates (ADCs) such as enfortumab vedotin (EV) and sacituzumab govitecan (SG) are novel treatments increasingly used for metastatic urothelial carcinoma (mUC). SG is also utilized in breast cancer (mBC). Radiotherapy (RT) palliation or consolidation of sites mUC bladder preservation localized disease. There limited data evaluating the safety ADCs when combined with RT. We sought to characterize RT patients mBC receiving EV SG. Methods: A bi-institutional...
105 Background: Next-generation sequencing (NGS) testing is used for prognostication and to guide treatment choice in metastatic prostate cancer (mPC). While NGS increasingly available, how its usage has impacted real-world selection of next line therapies including olaparib mPC unclear. Methods: We performed a retrospective observational study using the nationwide U.S. Flatiron Health electronic health record-derived de-identified database with patients diagnosed between January 1 st ,...
319 Background: BRCA mutations have emerged as critical biomarkers for guiding the use of PARP inhibitors in metastatic prostate cancer. Timely access to testing remains a challenge, especially those enrolled Medicare Advantage (MA) plans, which cover more than half beneficiaries but often impose greater restrictions Traditional (TM). This study compares timeliness patients with cancer (mPCa) across these two insurance types. Methods: A retrospective cohort was conducted using Flatiron...
320 Background: Next-Generation Sequencing (NGS) testing has become a cornerstone of personalized medicine in cancer care. With Medicare Advantage (MA) now covering nearly half all beneficiaries, understanding differences access to timely NGS between MA and Traditional (TM) is crucial, particularly given MA’s potential restrictions. This study aims evaluate the effectiveness plans on patients with metastatic prostate (mPCa). Methods: retrospective cohort was based Flatiron Health...
166 Background: As MA plans gain popularity, covering over half of eligible beneficiaries in 2023, concerns have arisen about their ability to manage complex cancer care due pre-authorization requirements and limited physician networks. The study compares Traditional Medicare (TM) delivering timely hormonal intensification, specifically through addition oral novel therapy (NHT) castration therapy, which became central part treating metastatic hormone-sensitive prostate (mHSPC) by 2018....
Importance: Artificial intelligence (AI) foundation models such as Segment Anything Model 2 (SAM 2) offer potential for semi-automated image segmentation with minimal fine-tuning, but their performance in specialized clinical tasks like radiation therapy planning are not well characterized. Objective: To evaluate the of SAM segmenting pre-operative intact prostate and post-operative fossa targets radiotherapy planning. Design, Setting, Participants: Retrospective cohort study deploying...
TPS4623 Background: Patients with muscle invasive bladder cancer (MIBC) may not be candidates for cisplatin-based chemotherapy based on their comorbidities and clinical status. Based EV-103 cohort H, patients localized, cisplatin ineligible MIBC respond well to enfortumab vedotin (EV), 36% pathologic complete responses (pCRs). Radiation (XRT) is also an effective therapy MIBC, recent retrospective data showing safety when combining XRT-EV. Therefore, we designed a trial EV XRT improve pCR...
Deep learning (DL) models have demonstrated state-of-the-art performance in the classification of diagnostic imaging oncology. However, DL for medical images can be compromised by adversarial images, where pixel values input are manipulated to deceive model. To address this limitation, our study investigates detectability oncology using multiple detection schemes. Experiments were conducted on thoracic computed tomography (CT) scans, mammography, and brain magnetic resonance (MRI). For each...