Articles tagged with Diagnostic Accuracy
A study published in Radiology found that radiologists and physicians who received local explanations from artificial intelligence (AI) systems performed better and made faster diagnoses than those who received global explanations. The researchers also found that AI advice was trusted more quickly when it provided local explanations.
A new study published in JAMA Network Open found that using Chat GPT Plus does not significantly improve the accuracy of doctors' diagnoses, but it outperformed conventional methods in certain cases. The researchers suggest that physicians need more training and experience with AI to capitalize on its potential.
A new study from Japan confirms that definitive criteria for lower cachexia prevalence rates are linked to reduced survival rates in cancer patients. The study found that the diagnostic criteria used for cachexia detection can affect reported prevalence and survival outcomes, emphasizing the need for accurate diagnosis.
Radiologists who had access to questionnaire information during MRI interpretation achieved a near statistically perfect agreement with spine specialists. This improved diagnostic accuracy and confidence, avoiding diagnostic discrepancies between radiologists and spine specialists.
A new study finds that AI-powered models exhibit similar levels of accuracy as ophthalmologists in identifying infectious keratitis, a leading cause of corneal blindness worldwide. The AI models displayed a sensitivity and specificity of 89.2% and 93.2%, respectively, matching the diagnostic accuracy of human experts.
Researchers used fluorescein angiography to visualize neural blood flow in rats and rabbits with chronic nerve compression neuropathy, correlating findings with electrodiagnostic testing. The study showed promising results for the method's potential to improve diagnosis and treatment outcomes in carpal tunnel surgery.
A study by Osaka Metropolitan University found that ChatGPT's diagnostic performance for brain tumors was comparable to that of neuroradiologists, with an accuracy rate of 73%. The model's performance varied depending on the type of clinical report written, with higher accuracy when using reports from neuroradiologist writers.
A systematic review of risk stratification tools found high-sensitivity troponin tests to have the highest diagnostic accuracy for ruling out acute coronary syndrome. However, these findings require further validation before recommending their use in primary care settings.
A Kobe University study found that the widely used casual blood glucose testing method for screening pregnant women with gestational diabetes mellitus misses 70% of cases. The researchers recommend using more reliable tests to prevent pregnancy complications and type II diabetes in mothers and children.
A study published in JAMA Network Open found that white patients are more likely to receive related diagnostic testing than Black patients after being discharged from the emergency department. This disparity raises concerns about potential overuse of tests in white patients versus undertesting and missed diagnoses in Black patients.
A recent study by WVU researchers found that smartwatches and clinical testing measures for heart rate variability have significant differences. The study used simulation analysis to determine the validity of wearable devices' calculations, revealing that some methods produce wider ranges of error in measurement.
A new imaging device that combines optical coherence tomography (OCT) with traditional otoscopy improves diagnostic capabilities for hearing clinics. The integrated device provides detailed views of the eardrum and middle ear, enabling more accurate diagnoses and treatment.
Researchers at Osaka Metropolitan University found that AI model GPT-4 outperformed a weaker version of the model and was on par with radiology residents in diagnosing musculoskeletal conditions. However, it struggled to match the accuracy of board-certified radiologists.
Researchers developed an approach to clearly separate physical and psychosocial components of pain, allowing for more targeted treatment. The new method combines measuring body signals, self-disclosure, and computerized evaluation to create two indices: one for physical component and one for psychosocial component.
Researchers found large language models are more accurate with concise, textbook-like medical questions than patient-written summaries. The models achieved higher accuracy when using standardized language, but struggled with variable phrasing and format of patient write-ups.
A new study found that hospital pneumonia diagnoses are often uncertain and revised, with over half of all cases involving a change in diagnosis. This uncertainty can lead to poorer health outcomes for patients who initially lack a pneumonia diagnosis but later receive a diagnosis.
A novel AI tool has been shown to accurately estimate gestational age from blind ultrasound sweeps, comparable to expert sonographers. The technology has the potential to democratize prenatal care in resource-limited settings by expanding access to quality diagnostic tools without the need for expensive equipment or specialized training.
Researchers found high diagnostic accuracy for identifying Alzheimer's disease using amyloid probability score 2 blood test and p-tau217 levels. The study suggests blood tests could influence clinical care, warranting further evaluation.
A study by University of Washington School of Medicine found that a cognitive test is an inaccurate predictor of athletes' concussions. Instead, symptom reports were the most accurate indicator of concussion. The study involved 92 NCAA Division I athletes who sustained a concussion and their teammates as matched control subjects.
Researchers at Queen Mary University of London developed a predictive test using fMRI scans that detects changes in the brain's 'default mode network' to predict dementia. The test showed accurate predictions up to nine years before an official diagnosis, with greater than 80% accuracy and a two-year margin of error.
The study uses the Dermatological Vision Dataset and compares vision transformers with traditional CNNs, achieving an impressive 97.8% accuracy on validation sets. Integrating ViTs into dermatology represents a promising step toward more accurate diagnostics.
The Royal Statistical Society has published new recommendations to improve the evaluation of diagnostic tests, including guidelines on study design, regulation, and transparency. The report aims to prevent future issues with diagnostic test evaluations, particularly those related to infectious diseases.
Researchers identified a unique autoantibody signature in approximately 10% of patients with MS, appearing years before symptom onset. This finding could revolutionize patient care and treatment strategies for multiple sclerosis.
A new lab-on-a-chip system combines optofluidics and nanopores to rapidly test for SARS-CoV-2 and Zika viruses with high accuracy. The tool, developed by UC Santa Cruz researchers, can detect viruses at extremely low concentrations and outperform PCR tests in some cases.
Researchers have developed an AI-driven test that accurately diagnoses ovarian cancer in women clinically classified as normal, improving detection of early-stage disease. The test uses machine learning and blood metabolite information to assign a probability of disease presence or absence, offering a more clinically informative approach.
Diagnostic errors in hospitalized adults who died or were transferred to the intensive care unit were common, with problems in test choice and clinician assessment identified as high-priority areas for improvement. The study analyzed 2,428 patient records at 29 hospitals and found that these errors led to significant patient harm.
A multicenter randomized clinical vignette survey study found that AI models, when systematically biased, reduced diagnostic accuracy. The study involved hospitalist physicians, nurse practitioners, and physician assistants from 13 states.
Researchers developed a fully automated solution using knowledge engineering methods to aggregate independent diagnoses and increased diagnostic accuracy from 46% to 76%. The collective solution improved across various medical specialties, chief complaints, and diagnosticians' tenure levels.
A study by Medical University of Vienna found that AI algorithms in smartphone applications perform well in diagnosing pigmented skin lesions, but are outperformed by doctors in treatment decisions. The 7-class algorithm showed equivalent diagnostic accuracy to experts, while the ISIC algorithm performed worse.
Researchers at the University of Massachusetts Amherst have developed a new method for DNA detection that is 100 times more sensitive than traditional methods. This breakthrough enables fast and accurate disease diagnosis, reducing wait times for lab processing from days to minutes.
A pilot study found that ChatGPT performed as well as doctors in suggesting the most likely diagnoses for patients being assessed in emergency medicine departments. The chatbot generated a list of likely diagnoses and suggested the most likely option, with a large overlap between its shortlist and those of doctors.
A radiomic-based model using T2-weighted MRI data achieved high accuracy in diagnosing pediatric Crohn disease, outperforming expert radiologists. The model was ensembled with clinical data to further improve performance.
A new study found that information-sharing networks among clinicians improve patient care, especially for the worst-performing doctors. The researchers created an app that connected doctors in anonymous networks, showing their peers' diagnostic risk estimates and treatment recommendations.
A new position paper reviews TBS for fracture risk prediction, treatment initiation and monitoring in osteoporosis. The evidence-based guide provides practical guidance for clinicians to integrate TBS into clinical practice.
The American Phytopathological Society has published a focus issue on critical biosecurity gaps in US plant disease diagnostics, highlighting the need for harmonized diagnostics within the agricultural biosecurity system. The focus issue addresses assay validation methods, including high-throughput screening and PCR/RPA techniques.
Columbia University researchers have created a novel, fully organic bioelectronic device that can acquire and transmit neurophysiologic brain signals while providing power to the implanted device. The device features a tiny transistor and has demonstrated high electrical performance, long-term stability, and biocompatibility.
A study published in Radiology found that AI algorithms with high diagnostic accuracy improved radiologists' detection of lung cancers on chest X-rays. High-accuracy AI led to more frequent changes in reader determinations, suggesting increased human trust in AI.
A new device using dielectrophoresis selectively isolates Mycobacterium tuberculosis from sputum samples, providing a purified sample for molecular confirmation. The technology has shown promising results in detecting TB in underserved areas, with high concordance rates compared to culture diagnosis.
Recent studies in the New Journal of Pharmaceutical Analysis feature novel diagnostic tools, RNA sequencing-based workflows, and mechanical property evaluations to enhance cancer and cardiovascular disease treatment outcomes. These innovations aim to improve the therapeutic effect of drugs and promote personalized medicine.
A new method called 5PSeq has been developed to quickly assess bacterial response to antibiotics, with potential implications for treating antibiotic-resistant infections. The method measures mRNA translation and decay, revealing how bacteria interact with environmental factors and stressors.
Researchers developed a 'Skeletal Age' metric to assess impact of fractures on mortality, revealing a loss of one to seven years of life depending on gender, age, and bone site. The online calculator measures bone fragility to help doctors and patients understand fracture risks.
A new laser-based breathalyzer using artificial intelligence can detect COVID-19 in real-time with excellent accuracy. The technology, powered by frequency comb spectroscopy and machine learning, analyzes the unique chemical fingerprint of each breath sample to identify specific health conditions.
A diagnostic study using 4,095 retinal fundus images found that biomarker-based AI algorithms can be susceptible to racial bias, even when trained on raw images. This issue highlights the need for careful evaluation of AI performance in diverse populations.
Scientists have successfully regulated the flow of single molecules in a solution by opening and closing a nanovalve, which could revolutionize chemical and biochemical synthesis. This technology has the potential to detect pathogens with high sensitivity and create new materials for various industries.
Researchers at Karolinska Institutet have discovered that a type of sugar molecule in blood is associated with the level of tau protein, a critical factor in severe dementia. Measuring blood glycan levels can predict Alzheimer's disease risk to 80% accuracy, almost a decade before symptoms appear.
A Swansea University study reveals the potential of machine learning in identifying Ankylosing Spondylitis (AS) patients, reducing diagnosis delays from eight years to earlier. The research uses a national data repository to develop a predictive model for AS detection, empowering GPs to refer patients more efficiently.
A new urine gene test can detect genetic mutations in urine that predict bladder cancer up to 12 years in advance of clinical symptoms. The test was trialled on over 50,000 participants and showed promising results, suggesting its potential for early detection and reduced unnecessary invasive procedures.
A new method developed by Flatiron Institute researchers can diagnose COVID-19 with near-perfect accuracy, even in asymptomatic patients. The technique monitors the body's molecular response to a viral attack and measures mRNA molecules to identify an immune response.
Researchers at Aston University have discovered a new approach to process LDF light signals, allowing for more precise measurement of blood flow in specific areas of the vascular bed. This innovation has shown significant improvement in diagnostic accuracy for detecting microvascular changes in patients with type 2 diabetes and age-spe...
Researchers at the University of Wisconsin-Madison have developed a machine-learning model that detects cancers at an early stage by analyzing fragments of cell-free DNA in plasma. The technique, which uses readily available lab materials, distinguished people with any stage of cancer from healthy individuals 91% of the time.
A team of researchers from Chung-Ang University evaluated three different MALDI-TOF MS approaches used in domestic clinical settings for the identification of molds. They compared the performance and diagnostic accuracy of the Bruker Biotyper, ASTA MicroIDSys, and Vitek MS instruments.
A new AI tool developed by Brigham and Women's Hospital improves the accuracy of time-critical pathology diagnostics during surgery. The tool, which leverages deep-learning technology, translates frozen tissue samples into high-quality images, increasing diagnostic accuracy and reducing the need for lengthy laboratory tests.
Researchers developed an AI-based neural network to detect early knee osteoarthritis from x-ray images, matching doctors' diagnoses in 87% of cases. This method could help reduce unnecessary examinations, treatments, and even knee joint replacement surgery.
Physicians in Brazil's states of São Paulo and Maranhão used telemedicine for various purposes, including staff meetings, training sessions, and patient consultations. However, the technology has limitations and requires regulation to ensure quality care.
The study analyzed over 180,000 patients, finding that depression, addiction, and stress reaction disorder are the most uncertain diagnoses. Doctors can use this data to predict treatment outcomes and improve care for patients, ultimately leading to better evidence-based follow-up.
Researchers developed an AI model to analyze spatial and temporal gait parameters, identifying key features for diagnosing Parkinson's. The model achieved high diagnostic accuracy, reducing the probability of error in clinical assessments.
A study published in the Journal of Alzheimer's Disease found that analyzing patients' drawing tests can help diagnose Alzheimer's disease (AD) and distinguish it from dementia with Lewy bodies (DLB). The researchers discovered that DLB patients showed differences in speed-, pressure-, and pause-related features, whereas AD patients sh...
Researchers developed a machine learning-based diagnostic method that combines genomic sequencing and analysis of patients' immune response for remarkable accuracy. The approach identifies and predicts sepsis cases with high accuracy, potentially exceeding current diagnostic capabilities.
A systematic review and meta-analysis found that pulse oximetry has a high false negative rate, making it inaccurate for triaging patients with suspected carbon monoxide poisoning. The researchers recommend developing alternative methods for rapid screening of carbon monoxide levels in capillaries.
A study published in the Canadian Journal of Cardiology found that smartwatch health apps detecting atrial fibrillation generated a high rate of false positives and inconclusive results, especially in patients with certain cardiac conditions. Better algorithms and machine learning may help improve the accuracy of these devices.