Loading...
Loading...
Medical imaging (X-ray, CT, MRI), electronic health records, clinical trials, ECG/EEG, pathology
14,759 datasets
Radiographic techniques were employed to determine the relative dispositions of the bony pelvic structure and of the seat belt. This investigation of anatomical aspects of lap strap fit was published by the NSW Government. The report concludes that lap strap fit can be assessed from surface anatomy by a simple technique and suggests revisions to seat belt geometry specifications.
A 2026 study by Marc Leon evaluates five large language models (O1, O3-mini-high, DeepSeek-R1, GPT-4, Llama3-OpenBioLLM-70B) on 15 high-fidelity cardiac surgery reasoning tasks. The dataset contains normalized performance scores across 10 evaluation dimensions, including scenario comprehension, patient safety, and hallucination avoidance, from a blinded two-phase evaluation by senior surgeons. It also records rating shifts between evaluation rounds, showing a 7.57% revision rate from affirmative to negative.
A 2026 study by Marc Leon presents a two-phase evaluation of five large language models (O1, O3-mini-high, DeepSeek-R1, GPT-4, Llama3-OpenBioLLM-70B) on 15 expert-curated cardiac surgery scenarios. The dataset contains normalized performance scores across 10 weighted evaluation dimensions, including scenario comprehension, patient safety, and hallucination avoidance. It also documents rating shifts from a blinded evaluation by senior surgeons, revealing patterns of human-AI collaboration.
A research paper presents a two-phase evaluation framework for large language models in cardiac surgery. The study includes 15 high-fidelity clinical scenarios developed by senior surgeons and evaluates five LLMs using a 10-dimensional weighted framework. The dataset, a 369.8 KB PDF, was authored by Marc Leon and last updated on 2026-05-29.
A 2026 study by Marc Leon presents a blinded two-phase evaluation of five large language models on 15 high-fidelity cardiac surgery scenarios. The dataset contains normalized performance scores for models including O1, O3-mini-high, DeepSeek-R1, GPT-4, and Llama3-OpenBioLLM-70B across 10 weighted evaluation dimensions. Results show performance variation and highlight a collaboration imbalance where clinicians over-accepted incorrect model reasoning.
Marc Leon's dataset contains results from a blinded two-phase evaluation of five large language models on 15 high-fidelity cardiac surgery scenarios. The data includes normalized performance scores across 10 weighted evaluation dimensions and records of rating revisions by senior surgeons. The dataset was last updated on 2026-05-29 and is licensed under CC-BY-4.0.
A blinded two-phase evaluation of five large language models on 15 high-fidelity cardiac surgery reasoning tasks. The dataset includes normalized performance scores across 10 weighted evaluation dimensions and records of rating revisions by senior surgeons. It was authored by Marc Leon and last updated in May 2026.
A blinded two-phase evaluation of five large language models on 15 high-fidelity cardiac surgery reasoning tasks. The dataset contains normalized performance scores across 10 weighted evaluation dimensions, including scenario comprehension and patient safety, and tracks rating revisions by senior surgeons. It was authored by Marc Leon and last updated in May 2026.
15 high-fidelity cardiac surgery scenarios were used to evaluate five large language models on a 10-dimensional weighted framework. Median normalized scores ranged from 0.521 for Llama3-OpenBioLLM-70B to 0.896 for O1, with scenario comprehension scoring highest and patient safety lowest. The dataset, created by Marc Leon and published on figshare in 2026, captures a blinded two-phase evaluation where surgeons revised 7.57% of ratings from affirmative to negative after seeing reference answers.
Five large language models were evaluated on 15 high-fidelity cardiac surgery scenarios by senior surgeons using a 10-dimensional weighted framework. Median normalized scores ranged from 0.521 for Llama3-OpenBioLLM-70B to 0.896 for O1, with scenario comprehension scoring highest and patient safety lowest. The dataset, created by Marc Leon and last updated in May 2026, captures model performance and evaluator judgment shifts in a blinded two-phase study.
A blinded two-phase evaluation assessed five large language models on 15 high-fidelity cardiac surgery reasoning tasks. Median normalized scores ranged from 0.521 to 0.896, with O1 achieving the highest score. The study, authored by Marc Leon and updated in May 2026, found that overacceptance of incorrect AI reasoning was a dominant collaboration imbalance.
Five large language models were evaluated on 15 high-fidelity cardiac surgery scenarios by senior surgeons. O1 achieved the highest median normalized score (0.896), while patient safety and hallucination avoidance were the lowest-scoring dimensions across models. The dataset, authored by Marc Leon and last updated in May 2026, documents the evaluation framework and results, concluding that LLMs are not yet ready for safe use in complex surgical settings.
A retrospective study of 133 patients with chronic Chagas cardiomyopathy who received implantable cardioverter-defibrillators, primarily for secondary prevention. The dataset includes demographic, clinical, laboratory, Rassi score, and defibrillation threshold test data, with a mean clinical follow-up of 1728 days. The research was authored by Marco Paulo Cunha Campos and published via paperswithcode.
A six-month retrospective study of 42 patients on online hemodiafiltration who switched to sucroferric oxyhydroxide. The data, collected by Aníbal Ferreira of Hospital Curry Cabral, tracks monthly serum phosphorus levels, pill burden, and intravenous iron medication. It shows a 67% reduction in prescribed pills per day after the treatment switch.
86 code blue (cardiac arrest) cases from 2022 at a tertiary care hospital in Mangalore, India, were analyzed in this retrospective record-based study. The data was collected from patient files by author J. Radhakrishnan and analyzed with descriptive and inferential statistical methods. The study reports patient demographics, response times, CPR initiation, and survival outcomes.
105 children aged 5 years or more diagnosed with congenital hypothyroidism were studied via parent/caregiver questionnaires and medical records. 72.4% of subjects showed symptoms related to vestibulocochlear disorder, with dizziness/vertigo at 56.2%, hearing loss at 43.8%, and tinnitus at 12.4%. The pilot study by Caio Leônidas De Andrade found statistical correlations between symptoms and clinical factors like neonatal screening age.
49 eyes from 37 patients aged 10 to 50 years were analyzed in a 2017 retrospective study at the Instituto Panamericano da Visão in Goiânia, Brazil. The research evaluated the safety and efficacy of Transepithelial Crosslinking (CXL) for progressive keratoconus, measuring visual acuity, astigmatism, and pachymetry over 12 months. The dataset was authored by Roseane Lucena Marquez and is available via paperswithcode.
A cross-sectional observational study collected from medical records at a University Hospital in the State of São Paulo. The dataset likely contains patient profiles and prophylaxis administration rates for venous thromboembolism (VTE). The study, authored by Arthur Curtarelli, found an overall correct prophylaxis prescription rate of 42.1%.
113 cancer patients experienced immune-related adverse events in a cohort of 297 patients receiving immune checkpoint inhibitor combination therapy. Yao Qiu published this clinical study on figshare in 2026, analyzing incidence, severity, onset time, and risk factors for adverse events. The findings suggest corticosteroid use is a protective factor and combination with targeted therapy increases risk for endocrine adverse events.
28,003 patients with ductal carcinoma in situ and 275,836 patients with stage I-III invasive breast cancer were identified from the Netherlands Cancer Registry between 1989 and 2017. This data record supports a study comparing the risk of developing contralateral breast cancer between these patient groups, accounting for factors like age and treatment. The underlying data for figures is available via figshare, while the full dataset is accessible upon request from the registry.