Significant Difference In Accuracy Research Articles

Evaluation of ChatGPT for Patient Counseling in Kidney Stone Clinic: A Prospective Study.

Introduction: The potential of large language models (LLMs) is to improve the clinical workflow and to make patient care efficient. We prospectively evaluated the performance of the LLM ChatGPT as a patient counseling tool in the urology stone clinic and validated the generated responses with those of urologists. Methods: We collected 61 questions from 12 kidney stone patients and prompted those to ChatGPT and a panel of experienced urologists (Level 1). Subsequently, the blinded responses of urologists and ChatGPT were presented to two expert urologists (Level 2) for comparative evaluation on preset domains: accuracy, relevance, empathy, completeness, and practicality. All responses were rated on a Likert scale of 1 to 10 for psychometric response evaluation. The mean difference in the scores given by the urologists (Level 2) was analyzed and interrater reliability (IRR) for the level of agreement in the responses between the urologists (Level 2) was analyzed by Cohen's kappa. Results: The mean differences in average scores between the responses from ChatGPT and urologists showed significant differences in accuracy (p < 0.001), empathy (p < 0.001), completeness (p < 0.001), and practicality (p < 0.001), except for the relevance domain (p = 0.051), with ChatGPT's responses being rated higher. The IRR analysis revealed significant agreement only in the empathy domain [k = 0.163, (0.059-0.266)]. Conclusion: We believe the introduction of ChatGPT in the clinical workflow could further optimize the information provided to patients in a busy stone clinic. In this preliminary study, ChatGPT supplemented the answers provided by the urologists, adding value to the conversation. However, in its current state, it is still not ready to be a direct source of authentic information for patients. We recommend its use as a source to build a comprehensive Frequently Asked Questions bank as a prelude to developing an LLM Chatbot for patient counseling.

Enhancing systematic reviews in orthodontics: a comparative examination of GPT-3.5 and GPT-4 for generating PICO-based queries with tailored prompts and configurations.

The rapid advancement of Large Language Models (LLMs) has prompted an exploration of their efficacy in generating PICO-based (Patient, Intervention, Comparison, Outcome) queries, especially in the field of orthodontics. This study aimed to assess the usability of Large Language Models (LLMs), in aiding systematic review processes, with a specific focus on comparing the performance of ChatGPT 3.5 and ChatGPT 4 using a specialized prompt tailored for orthodontics. Five databases were perused to curate a sample of 77 systematic reviews and meta-analyses published between 2016 and 2021. Utilizing prompt engineering techniques, the LLMs were directed to formulate PICO questions, Boolean queries, and relevant keywords. The outputs were subsequently evaluated for accuracy and consistency by independent researchers using three-point and six-point Likert scales. Furthermore, the PICO records of 41 studies, which were compatible with the PROSPERO records, were compared with the responses provided by the models. ChatGPT 3.5 and 4 showcased a consistent ability to craft PICO-based queries. Statistically significant differences in accuracy were observed in specific categories, with GPT-4 often outperforming GPT-3.5. The study's test set might not encapsulate the full range of LLM application scenarios. Emphasis on specific question types may also not reflect the complete capabilities of the models. Both ChatGPT 3.5 and 4 can be pivotal tools for generating PICO-driven queries in orthodontics when optimally configured. However, the precision required in medical research necessitates a judicious and critical evaluation of LLM-generated outputs, advocating for a circumspect integration into scientific investigations.

Is ChatGPT a trusted source of information for total hip and knee arthroplasty patients?

While internet search engines have been the primary information source for patients' questions, artificial intelligence large language models like ChatGPT are trending towards becoming the new primary source. The purpose of this study was to determine if ChatGPT can answer patient questions about total hip (THA) and knee arthroplasty (TKA) with consistent accuracy, comprehensiveness, and easy readability. We posed the 20 most Google-searched questions about THA and TKA, plus ten additional postoperative questions, to ChatGPT. Each question was asked twice to evaluate for consistency in quality. Following each response, we responded with, "Please explain so it is easier to understand," to evaluate ChatGPT's ability to reduce response reading grade level, measured as Flesch-Kincaid Grade Level (FKGL). Five resident physicians rated the 120 responses on 1 to 5 accuracy and comprehensiveness scales. Additionally, they answered a "yes" or "no" question regarding acceptability. Mean scores were calculated for each question, and responses were deemed acceptable if ≥ four raters answered "yes." The mean accuracy and comprehensiveness scores were 4.26 (95% confidence interval (CI) 4.19 to 4.33) and 3.79 (95% CI 3.69 to 3.89), respectively. Out of all the responses, 59.2% (71/120; 95% CI 50.0% to 67.7%) were acceptable. ChatGPT was consistent when asked the same question twice, giving no significant difference in accuracy (t = 0.821; p = 0.415), comprehensiveness (t = 1.387; p = 0.171), acceptability (χ2 = 1.832; p = 0.176), and FKGL (t = 0.264; p = 0.793). There was a significantly lower FKGL (t = 2.204; p = 0.029) for easier responses (11.14; 95% CI 10.57 to 11.71) than original responses (12.15; 95% CI 11.45 to 12.85). ChatGPT answered THA and TKA patient questions with accuracy comparable to previous reports of websites, with adequate comprehensiveness, but with limited acceptability as the sole information source. ChatGPT has potential for answering patient questions about THA and TKA, but needs improvement.

Left-handed voices? Examining the perceptual learning of novel person characteristics from the voice.

We regularly form impressions of who a person is from their voice, such that we can readily categorise people as being female or male, child or adult, trustworthy or not, and can furthermore recognise who specifically is speaking. How we establish mental representations for such categories of person characteristics has, however, only been explored in detail for voice identity learning. In a series of experiments, we therefore set out to examine whether and how listeners can learn to recognise a novel person characteristic. We specifically asked how diagnostic acoustic properties underpinning category distinctions inform perceptual judgements. We manipulated recordings of voices to create acoustic signatures for a person's handedness (left-handed vs. right-handed) in their voice. After training, we found that listeners were able to successfully learn to recognise handedness from voices with above-chance accuracy, although no significant differences in accuracy between the different types of manipulation emerged. Listeners were, furthermore, sensitive to the specific distributions of acoustic properties that underpinned the category distinctions. We, however, also find evidence for perceptual biases that may reflect long-term prior exposure to how voices vary in naturalistic settings. These biases shape how listeners use acoustic information in the voices when forming representations for distinguishing handedness from voices. This study is thus a first step to examine how representations for novel person characteristics are established, outside of voice identity perception. We discuss our findings in light of theoretical accounts of voice perception and speculate about potential mechanisms that may underpin our results.

Comparison of force profiles from two musculoskeletal palpation methods: A methodological study.

The Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) protocol recommends a 5 s and 1 kg force dynamic palpation around the lateral condylar pole of the temporomandibular joint. However, the accuracy and precision of the generated force are not known. To assess and compare the force profiles generated from dynamic palpation manually and using a palpometer, based on the forces and time recommendations suggested by the DC/TMD protocol. Nineteen healthy adults applied forces of 0.5 kg, 1.0 kg and 2.0 kg on a calibrated force sensor in a circular motion within target times of 2 s and 5 s. Participants used their right index finger for manual palpation and a calibrated palpometer for device-assisted palpation. Ten repetitions of each target force at both target times were applied. Time taken to complete each application was recorded. Repeated measures analysis of variance was used for analysis of accuracy measured as the relative difference between targeted force and actual force values and precision measured as the coefficient of variation (CV) within the 10 repeated measurements. Accuracy was significantly lower (better) and precision higher (lower CV) with the palpometer than with manual palpation (p < .001). There were significant differences in accuracy and precision between the different forces but not palpation times. Most participants could not achieve the target times and tended to be faster, irrespective of the palpation method (p > .063). A palpometer is a more accurate and precise palpation method for dynamic force assessment compared to manual palpation; however, it remains difficult to standardize the palpation duration.

Sharing Science Made Simple: Exploring the Quality and Readability of Published Lay Summaries

Lay summaries exist to bridge the gap that separates the scientific community from the general public. To foster improved science communication, this study examined the overall quality and readability of published lay summaries across peer-reviewed journals. We obtained 200 lay summaries published in four science journals: eLife, PLOS Medicine, Proceedings of the National Academy of Science (PNAS), and the Journal of Hepatology. Over 900 students across three semesters participated as raters of each summary using a rubric developed to assess the overall quality, accuracy, and accessibility of lay summaries across these journals. The Flesch Reading Ease formula was used to determine the readability of the highest and lowest scoring summaries from each journal. eLife and the Journal of Hepatology had the highest and lowest mean scores for overall quality of 15.6 and 11.7 out of 20, respectively. There were statistically significant differences in accuracy and accessibility found across all journals (p&lt;0.0001). eLife had the highest scoring lay summary for readability. The differences in and lack of consistent scoring across journals with the rubric indicate that deficits exist in the overall quality and readability of published lay summaries. These findings may support the development of guidelines that incorporate elements of the rubric used to write effective lay summaries.

Construction of Semantic Coherence Diagnosis Model of English Text based on Sentence Semantic Map

The current English composition automatic correction system rarely involves coherence quality analysis of compositions. Therefore, a semantic coherence diagnosis model for English texts was constructed based on sentence semantic maps, and its effectiveness was verified through experiments. Experimental results show that sub Graphical model 10, 12 and 13 exceed 200 on the first two test texts and 1 on the last two test texts. However, the differences between these three subgraphs on different test texts are not significant, with differences below 30 and 0.3. In addition, when extracting incoherent sentences, the F1 value reaches the optimal value at a threshold of 0.34, which is 87.54%. When the threshold is fixed at 0.34, the accuracy of extracting non coherent sentences also increases with the number of articles, reaching a maximum of 88.43%. At the same time, there was no significant difference in accuracy, recall, and F1 values among different English composition numbers, maintaining between 83% and 89%. The Pearson coefficient calculated in the actual comparison with the teacher's manual composition score is 0.6025, indicating a strong correlation between the two, indicating that the diagnostic results are reliable. Overall, the diagnostic model constructed in the study has strong accuracy and effectiveness, and is practical in the diagnosis of semantic coherence in actual English texts.

Developing, validating, and comparing an analytical method to simultaneously detect z-drugs in urine samples using the QuEChERS approach with both liquid chromatography and gas chromatography-tandem mass spectrometry

Detecting z-drugs, a sedative-hypnotic medication, is also misused for criminal activities. Therefore, the analysis of urine samples is crucial for clinical and forensic purposes. We conducted a study where we developed, validated, and compared an analytical method for simultaneously detecting z-drugs in urine samples. Our approach uses the QuEChERS method for sample preparation, combined with liquid chromatography (LC) and gas chromatography (GC) coupled with tandem mass spectrometry (MS/MS). We optimized the QuEChERS method to effectively extract z-drugs from urine samples while minimizing matrix effects and achieving high recovery rates. After extraction, we split the samples into two parts for analysis using LC-MS/MS and GC–MS/MS. We validated our methods, and the results showed good linearity over a broad concentration range (1–200 ng/mL) for each z-drug. The limits of detection and quantification were within clinically relevant ranges, ensuring sensitivity for detecting z-drugs in urine samples. We compared the two chromatographic techniques by analyzing a set of urine samples spiked with known concentrations of z-drugs using both LC-MS/MS and GC–MS/MS methods and then applied to the real samples. The results were statistically analyzed to assess any significant differences in accuracy and precision above 95 %, and both methods offered reliable and consistent results with the samples as well. In conclusion, our analytical method coupled with both LC-MS/MS and GC–MS/MS using the QuEChERS approach provides a comprehensive and robust solution for the simultaneous detection of z-drugs in urine samples. The choice between the two chromatographic techniques can be based on the specific z-drugs of interest and the required analytical performance. This method holds promise for applications in clinical toxicology, forensic analysis, and monitoring z-drug usage.

The Accuracy of the NSQIP Universal Surgical Risk Calculator Compared to Operation-Specific Calculators.

To compare the performance of the ACS NSQIP "universal" risk calculator (N-RC) to operation-specific RCs. Resources have been directed toward building operation-specific RCs because of an implicit belief that they would provide more accurate risk estimates than the N-RC. However, operation-specific calculators may not provide sufficient improvements in accuracy to justify the costs in development, maintenance, and access. For the N-RC, a cohort of 5,020,713 NSQIP patient records were randomly divided into 80% for machine learning algorithm training and 20% for validation. Operation-specific risk calculators (OS-RC) and OS-RCs with operation-specific predictors (OSP-RC) were independently developed for each of 6 operative groups (colectomy, whipple pancreatectomy, thyroidectomy, abdominal aortic aneurysm (open), hysterectomy/myomectomy, and total knee arthroplasty) and 14 outcomes using the same 80%/20% rule applied to the appropriate subsets of the 5M records. Predictive accuracy was evaluated using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC), and Hosmer-Lemeshow (H-L) P values, for 13 binary outcomes, and mean squared error for the length of stay outcome. The N-RC was found to have greater AUROC (P = 0.002) and greater AUPRC (P < 0.001) compared to the OS-RC. No other statistically significant differences in accuracy, across the 3 risk calculator types, were found. There was an inverse relationship between the operation group sample size and magnitude of the difference in AUROC (r = -0.278; P = 0.014) and in AUPRC (r = -0.425; P < 0.001) between N-RC and OS-RC. The smaller the sample size, the greater the superiority of the N-RC. While operation-specific RCs might be assumed to have advantages over a universal RC, their reliance on smaller datasets may reduce their ability to accurately estimate predictor effects. In the present study, this tradeoff between operation specificity and accuracy, in estimating the effects of predictor variables, favors the N-R, though the clinical impact is likely to be negligible.

The effect of instrumental and vocal musical experience on tone categorical perception in individuals in a Chinese dialect area

IntroductionPrevious studies have shown that music training modulates adults’ categorical perception of Mandarin tones. However, the effect of music training on tone categorical perception ability in individuals in Chinese dialect areas remains unclear.MethodsWe recruited three groups of college students majoring in music in Chongqing, a dialect area in southwestern China. These groups included an instrumental music group (n = 31), a vocal music group (n = 32), and an instrumental-vocalist music group (n = 30). Additionally, we recruited a group of college students who did not receive any music training (n = 32). The accuracy and reaction time of the four groups were measured using the oddball task and compared to explore the differences in their tone categorical perception ability.ResultsConsidering both between-tone category and within-tone category, the three music groups showed both greater accuracy and shorter reaction times than the non-music group. Regarding the three music groups, no significant differences in accuracy and reaction time were detected between the instrumental and vocal groups. However, the instrumental vocalist group outperformed both the instrumental and vocal groups in terms of accuracy and reaction time.DiscussionResults showed that music training seems to have a positive effect on the categorical perception of Mandarin tone categories in Chinese dialect areas, and demonstrated that the combination of instrumental and vocal music training can further enhance tone categorical perception. To some extent, these findings provide a theoretical basis for the improvement of individual tone perception ability in dialect areas, and provided theoretical support for music and language education.

Ionophore-Based Polymeric Sensors for Potentiometric Assay of the Anticancer Drug Gemcitabine in Pharmaceutical Formulation: A Comparative Study.

Gemcitabine is a chemotherapeutic agent used to treat various malignancies, including breast and bladder cancer. In the current study, three innovative selective gemcitabine hydrochloride sensors are developed using 4-tert-butylcalix-[8]-arene (sensor 1), β-cyclodextrin (sensor 2), and γ-cyclodextrin (sensor 3) as ionophores. The three sensors were prepared by incorporating the ionophores with o-nitrophenyl octyl ether as plasticizer and potassium tetrakis(4-chlorophenyl) borate as ionic additive into a polyvinyl chloride polymer matrix. These sensors are considered environmentally friendly systems in the analytical research. The linear responses of gemcitabine hydrochloride were in the concentration range of 6.0 × 10-6 to 1.0 × 10-2 mol L-1 and 9.0 × 10-6 to 1.0 × 10-2 mol L-1 and 8.0 × 10-6 to 1.0 × 10-2 mol L-1 for sensors 1, 2, and 3, respectively. Over the pH range of 6-9, fast-Nernst slopes of 52 ± 0.6, 56 ± 0.3, and 55 ± 0.8 mV/decade were found in the same order with correlation regressions of 0.998, 0.999, and 0.998, respectively. The lower limits of detection for the prepared sensors were 2.5 × 10-6, 2.2 × 10-6, and 2.7 × 10-6 mol L-1. The sensors showed high selectivity and sensitivity for gemcitabine. Validation of the sensors was carried out in accordance with the requirements established by the IUPAC, while being inexpensive and easy to use in drug formulation. A statistical analysis of the methods in comparison with the official method showed that there was no significant difference in accuracy or precision between them. It was shown that the new sensors could selectively and accurately find gemcitabine hydrochloride in bulk powder, pharmaceutical formulations, and quality control tests. The ionophore-based sensor shows several advantages over conventional PVC membrane sensor sensors regrading the lower limit of detection, and higher selectivity towards the target ion.

Effects of short-term exposure to moderate pure carbon dioxide levels on cognitive performance, health symptoms and perceived indoor environment quality

Educational buildings frequently experience elevated CO2 concentrations with inadequate ventilation and high occupancy, sometimes exceeding building guideline levels. Some studies reported detrimental impacts on cognitive performance of indoor CO2 levels, while others did not. To generate further evidence, we conducted an experiment in an environmentally controlled chamber. Sixty-nine healthy university students were exposed individually for 70 min, in three separate sessions, to three CO2 conditions of 600, 1500 and 2100 ppm (crossover design). With fixed ventilation rates, pure CO2 was injected to achieve different exposure levels. A validated neurobehavioral BARS test battery was used to assess participants’ cognitive performance. Participants gave subjective ratings of indoor environment and reported any health symptom through questionnaires. Comparing elevated CO2 levels to 600 ppm, after adjusting for potential confounders, results showed significant improved performance, that is, responses were quicker in two out of ten tests, and no significant differences in accuracy for any test. Under 1500 ppm, participants rated the air quality significantly higher than at 600 ppm, but there were no differences at 2100 ppm. Differences were not significant on thermal sensation, perceived lighting quality, perceived noise level, or health symptoms for comparisons between conditions. Results indicate no clear link between pure CO2 levels below 2100 ppm and cognitive performance, perceived indoor environment quality and health symptoms. The findings are consistent with some prior studies, indicating that pure CO2 below 2100 ppm implies no harm in adults and should not be treated as a potential indoor pollutant in higher educational environments.

Unveiling the ChatGPT phenomenon: Evaluating the consistency and accuracy of endodontic question answers.

Chatbot Generative Pre-trained Transformer (ChatGPT) is a generative artificial intelligence (AI) software based on large language models (LLMs), designed to simulate human conversations and generate novel content based on the training data it has been exposed to. The aim of this study was to evaluate the consistency and accuracy of ChatGPT-generated answers to clinical questions in endodontics, compared to answers provided by human experts. Ninety-one dichotomous (yes/no) questions were designed and categorized into three levels of difficulty. Twenty questions were randomly selected from each difficulty level. Sixty answers were generated by ChatGPT for each question. Two endodontic experts independently answered the 60 questions. Statistical analysis was performed using the SPSS program to calculate the consistency and accuracy of the answers generated by ChatGPT compared to the experts. Confidence intervals (95%) and standard deviations were used to estimate variability. The answers generated by ChatGPT showed high consistency (85.44%). No significant differences in consistency were found based on question difficulty. In terms of answer accuracy, ChatGPT achieved an average accuracy of 57.33%. However, significant differences in accuracy were observed based on question difficulty, with lower accuracy for easier questions. Currently, ChatGPT is not capable of replacing dentists in clinical decision-making. As ChatGPT's performance improves through deep learning, it is expected to become more useful and effective in the field of endodontics. However, careful attention and ongoing evaluation are needed to ensure its accuracy, reliability and safety in endodontics.

A - 175 EEG Phase Can be Predicted with Similar Accuracy across Cognitive States after Accounting for Power and SNR.

Electroencephalogram (EEG) phase can capture neural oscillation dynamics. Accordingly, EEG phase is a potential target for neurofeedback and brain-computer interfaces (BCIs). However, rigorous tests of the generalizability of real-time EEG phase are needed, requiring accurate phase estimates. We examined how cognitive states affect EEG phase prediction accuracy in the parieto-occipital alpha band. We identified datasets through public repositories. After preprocessing, we used the Educated Temporal Prediction Algorithm (ETP) to predict future peaks. We compared these predictions to the original signal, and defined accuracy from 0 to 100%. Our independent variable was cognitive state (eyes-open rest, eyes-closed rest, task), dependent variable was accuracy, and covariates were EEG instantaneous power and signal-to-noise ratio (SNR). We used a linear mixed-effects model, with random effects for dataset and individual. Across the 11 datasets, we had 543 participants and 1,641,074 predictions. There were no significant accuracy differences among the conditions (p < 1e-5), with a baseline accuracy of 59.07%. Power had an effect on accuracy, with a unit increase leading to a 13.1% increase in accuracy (p < 1e-5). SNR had an effect on accuracy, with an effect size of 0.46% (p < 1e-5), with a significant negative interaction with power with an effect size of -0.51% (p < 1e-5). Our results indicated that we could predict EEG phase accurately across cognitive conditions and datasets, with higher accuracy for high instantaneous band-power and SNR. Accordingly, real-time EEG phase experiments, closed-loop technologies, and BCIs should minimize external unwanted noise while targeting periods of high power, as opposed to manipulating experimental and cognitive conditions.

Accuracy and Retention of Molding Plates Used for Infants with Cleft Lip and Palate Fabricated from Different Materials: A Cross-Sectional Clinical Study.

The study aimed to comparatively assess the accuracy and retention of molding plates made of polyethylene terephthalate modified with glycol (PET-G) and polymethyl methacrylate (PMMA) materials, while also investigating the effect of adhesive addition on retention. A cross-sectional clinical study. The study included 30 infants diagnosed with non-syndromic cleft lip and palate (16 unilateral, 14 bilateral). Two molding plates were fabricated for each infant, and their accuracy and retention were evaluated. The data were analyzed using independent t-tests, Mann-Whitney U tests, and Wilcoxon rank tests. Accuracy: The accuracy of the molding plates was assessed by measuring the virtual gap between the inner surface of the plates and their working cast using Exocad software. The accuracy was evaluated in different regions (anterior, middle, and posterior) and compared between PET-G and PMMA materials. Retention: The retention of the molding plates was measured using a digital force gauge, which recorded the force required to dislodge the plates from the infant's mouth. The retention was compared between PET-G and PMMA materials, as well as the effect of adhesive addition on retention. Overall adaptation; PET-G plates showed a significantly smaller gap (mean= 0.264 ±0.106) compared to PMMA (mean= 0.362 ±0.130). Region-specific adaptation: PET-G plates demonstrated better accuracy in all regions (means of anterior =0.246, middle =0.262, posterior =0.282 µm). Significant differences in accuracy were observed in the middle and posterior regions compared to PMMA. Retention assessment revealed that the PET-G groups (with or without adhesive) exhibited significantly higher retention compared to the PMMA groups (P<0.01). PET-G plates demonstrated superior accuracy and retention compared to PMMA, with a significant difference observed in both accuracy and retention. Furthermore, the addition of denture adhesive had a positive effect on retention for both materials.

Assessment of environmental sound perception among listeners with hearing loss and cochlear implants

Perception of environmental sounds (PES), encompassing informationally and/or aesthetically salient nonspeech, nonmusical sounds, is crucial for safety, independence, and quality-of-life among listeners with hearing loss. The aims of this study were (1) to systematically review methodologies used to assess PES among cochlear implant (CI) users and identify performance trends in this population, and (2) to present preliminary data using a novel PES task requiring inference of materials and actions generating environmental sounds. For aim 1, PES in quiet using open- or closed-set response formats were most commonly used. PES accuracy in pediatric (3 studies) and adult (16 studies) CI users was highly variable but generally mediocre (mean correct: 31–87%). Most studies were cross-sectional; only two evaluated PES prospectively before and after CI. No significant differences in accuracy were reported between CI candidates and CI users. PES correlated in with measures of speech perception and spectro-temporal processing. For aim 2, we present preliminary findings on identification of materials and actions responsible for generating environmental sounds among normal hearing listeners and listeners with hearing loss using hearing aids or CIs. We discuss the types of materials and actions that are most difficult for listeners with hearing loss and common sources of confusion.

A Comparison of Retrospective ECG-Gated CT and Surgical or Angiographical Findings in Acute Aortic Syndrome.

The best cardiac phases in retrospective ECG-gated CT for detecting an intimal tear (IT) in aortic dissection (AD) and an ulcer-like projection (ULP) in an intramural hematoma (IMH) have not been established. This study aimed to compare the detection accuracy of diastolic-phase and systolic-phase ECG-gated CT for IT in AD and ULP in IMH, with subsequent surgical or angiographical confirmation as the reference standard.In total, 81 patients (67.6 ± 11.8 years; 41 men) who underwent emergency ECG-gated CT and subsequent open surgery or thoracic endovascular aortic repair for AD (n = 52) or IMH (n = 29) were included. The accuracies of detecting IT and ULP were compared among only diastolic-phase, only systolic-phase, and both diastolic-phase and systolic-phase methods of retrospective ECG-gated CT; surgical or angiographical findings were used as the reference standard. The detection accuracy for IT and ULP using only diastolic-phase, only systolic-phase, and both diastolic-phase and systolic-phase methods of ECG-gated CT was 93% [95% CI: 87-97], 94% [95% CI: 88-97], and 95% [95% CI: 90-97], respectively. There were no significant differences in detection accuracy among the 3 acquisition methods (P = 0.55). Similarly, there were no significant differences in the accuracy of detecting IT in AD (P = 0.55) and ULP in IMH (P > 0.99) among only diastolic-phase, only systolic-phase, and both diastolic- and systolic-phase ECG-gated CT.Retrospective ECG-gated CT for detecting IT in AD and ULP in IMH yields highly accurate findings. There were no significant differences seen among only diastolic-phase, only systolic-phase, and both diastolic-phase and systolic-phase ECG-gated CT.

Identification of foods in the breakfast and determination of nutritional facts using deep convolution neural networks

Food recognition plays a crucial role in various fields, including healthcare, nutrition, and the food industry. It involves identifying different types of foods or dishes from images, videos, or other data sources. In healthcare, food recognition aids individuals in monitoring their daily food intake and managing their diet. It also assists dietitians and nutritionists in creating personalized meal plans based on patients' nutritional requirements and preferences. This article focuses on the development of software that can recognize food products and predict their nutritional facts. The software extracts essential nutritional facts such as fat, carbohydrates, protein, and energy from the food products and compiles them into a comprehensive list. For each of the 20 food products, 36 food images were obtained, resulting in a total of 720 food images. To validate the accuracy of the trained models, six different images of each food product were set aside for external validation purposes. The rest of the images were then trained using deep learning algorithms, namely, GoogleNet, ResNet-50, and Inception-v3, in the MATLAB software. The training and validation processes yielded over 98% correct predictions for each of the deep learning algorithms. Although there were no significant differences in accuracy among the algorithms, GoogleNet stood out when considering both prediction accuracy and prediction time. The validated deep learning algorithms were employed in developing the software for food recognition and nutritional value determination. The results indicate that the developed software can reliably identify foods and provide their corresponding nutritional facts. This software holds significant potential for application in the nutrition and dietetic field and can be particularly useful in healthcare settings for monitoring the dietary intake of patients with chronic diseases such as diabetes, heart disease, or obesity. The system can track the types and quantities of foods consumed, offering personalized feedback to patients and healthcare providers.

Harmonization of supervised machine learning practices for efficient source attribution of Listeria monocytogenes based on genomic data

BackgroundGenomic data-based machine learning tools are promising for real-time surveillance activities performing source attribution of foodborne bacteria such as Listeria monocytogenes. Given the heterogeneity of machine learning practices, our aim was to identify those influencing the source prediction performance of the usual holdout method combined with the repeated k-fold cross-validation method.MethodsA large collection of 1 100 L. monocytogenes genomes with known sources was built according to several genomic metrics to ensure authenticity and completeness of genomic profiles. Based on these genomic profiles (i.e. 7-locus alleles, core alleles, accessory genes, core SNPs and pan kmers), we developed a versatile workflow assessing prediction performance of different combinations of training dataset splitting (i.e. 50, 60, 70, 80 and 90%), data preprocessing (i.e. with or without near-zero variance removal), and learning models (i.e. BLR, ERT, RF, SGB, SVM and XGB). The performance metrics included accuracy, Cohen’s kappa, F1-score, area under the curves from receiver operating characteristic curve, precision recall curve or precision recall gain curve, and execution time.ResultsThe testing average accuracies from accessory genes and pan kmers were significantly higher than accuracies from core alleles or SNPs. While the accuracies from 70 and 80% of training dataset splitting were not significantly different, those from 80% were significantly higher than the other tested proportions. The near-zero variance removal did not allow to produce results for 7-locus alleles, did not impact significantly the accuracy for core alleles, accessory genes and pan kmers, and decreased significantly accuracy for core SNPs. The SVM and XGB models did not present significant differences in accuracy between each other and reached significantly higher accuracies than BLR, SGB, ERT and RF, in this order of magnitude. However, the SVM model required more computing power than the XGB model, especially for high amount of descriptors such like core SNPs and pan kmers.ConclusionsIn addition to recommendations about machine learning practices for L. monocytogenes source attribution based on genomic data, the present study also provides a freely available workflow to solve other balanced or unbalanced multiclass phenotypes from binary and categorical genomic profiles of other microorganisms without source code modifications.

Classifying Winning Performances in International Women's Rugby Union.

The efficacy of isolated and relative performance indicators (PIs) has been compared in rugby union; the latter more effective at discerning match outcomes. However, this methodology has not been applied in women's rugby. The aim of this study was to identify PIs that maximize prediction accuracy of match outcome, from isolated and relative data sets, in women's rugby union. Twenty-six PIs were selected from 110 women's international rugby matches between 2017 and 2022 to form an isolated data set, with relative data sets determined by subtracting corresponding opposition PIs. Random forest classification was completed on both data sets, and feature selection and importance were used to simplify models and interpret key PIs. Models were used in prediction on the 2021 World Cup to evaluate performance on unseen data. The isolated full model correctly classified 75% of outcomes (CI, 65%-82%), whereas the relative full model correctly classified 78% (CI, 69%-86%). Reduced respective models correctly classified 74% (CI, 65%-82%) and 76% (CI, 67%-84%). Reduced models correctly predicted 100% and 96% of outcomes for isolated and relative test data sets, respectively. No significant difference in accuracy was found between data sets. In the relative reduced model, meters made, clean breaks, missed tackles, lineouts lost, carries, and kicks from hand were significant. Increased relative meters made, clean breaks, carries, and kicks from hand and decreased relative missed tackles and lineouts lost were associated with success. This information can be utilized to inform physical and tactical preparation and direct physiological studies in women's rugby.