Mean Intraclass Correlation Coefficient Research Articles

Quantification of water exchange across the blood-brain barrier using noncontrast MR fingerprinting.

A method is proposed to quantify cerebral blood volume ( ) and intravascular water residence time ( ) using MR fingerprinting (MRF), applied using a spoiled gradient echo sequence without the need for contrast agent. An in silico study optimized an acquisition protocol to maximize the sensitivity of the measurement to and changes. Its accuracy in the presence of variations in , , and was evaluated. The optimized protocol (scan time of 19 min) was then tested in a exploratory healthy volunteer study (10 volunteers, mean age 24 3, six males) at 3 T with a repeat scan taken after repositioning to allow estimation of repeatability. Simulations show that assuming literature values for and , no variation in , while fitting only and , leads to large errors in quantification of and , regardless of noise levels. However, simulations also show that matching , , , and , simultaneously is feasible at clinically achievable noise levels. Across the healthy volunteers, all parameter quantifications fell within the expected literature range. In addition, the maps show good agreement between hemispheres suggesting physiologically relevant information is being extracted. Expected differences between white and gray matter (p < 0.0001) and (p < 0.0001) are observed, and show no significant differences, p = 0.4 and p = 0.6, respectively. Moderate to excellent repeatability was seen between repeat scans: mean intra-class correlation coefficient of , , , and . We demonstrate that regional simultaneous quantification of , , , and using MRF is feasible in vivo.

Artificial intelligence to automate assessment of ocular and periocular measurements.

To develop and validate a deep learning facial landmark detection network to automate the assessment of periocular anthropometric measurements. Patients presenting to the ophthalmology clinic were prospectively enrolled and had their images taken using a standardised protocol. Facial landmarks were segmented on the images to enable calculation of marginal reflex distance (MRD) 1 and 2, palpebral fissure height (PFH), inner intercanthal distance (IICD), outer intercanthal distance (OICD), interpupillary distance (IPD) and horizontal palpebral aperture (HPA). These manual segmentations were used to train a machine learning algorithm to automatically detect facial landmarks and calculate these measurements. The main outcomes were the mean absolute error and intraclass correlation coefficient. A total of 958 eyes from 479 participants were included. The testing set consisted of 290 eyes from 145 patients. The AI algorithm demonstrated close agreement with human measurements, with mean absolute errors ranging from 0.22 mm for IPD to 0.88 mm for IICD. The intraclass correlation coefficients indicated excellent reliability (ICC > 0.90) for MRD1, MRD2, PFH, OICD, IICD, and IPD, while HPA showed good reliability (ICC 0.84). The landmark detection model was highly accurate and achieved a mean error rate of 0.51% and failure rate at 0.1 of 0%. The automated facial landmark detection network provided accurate and reliable periocular measurements. This may help increase the objectivity of periocular measurements in the clinic and may facilitate remote assessment of patients with tele-health.

Transfer accuracy of 3D printed versus CAD/CAM milled surgical guides for temporary orthodontic implants: a preclinical micro CT study

ObjectivesTemporary anchorage devices (TADs) have become an integral part of comprehensive orthodontic treatments. This study evaluated the transfer accuracy of three-dimensional (3D) printed and computer-aided design/computer-aided manufacturing (CAD/CAM) milled surgical guides for orthodontic TADs using micro-computed tomography (CT) imaging in a preclinical trial. MethodsOverall, 30 surgical guides were used to place TADs into typodonts; 3D printing and CAD/CAM milling were used to produce the guides. The virtual target positions of the TADs were compared to the real positions in terms of spatial and angular deviations using digital superimposition. Micro-CT imaging was used to detect the positions. To evaluate reliability, two investigators collected the measurements twice. Intra-rater and inter-rater correlations were tested. ResultsIn total, 60 palatal TADs evaluated. The mean coronal deviations in the print group ranged from 0.15 ± 0.20 mm to 0.71 ± 0.22 mm, whereas in the mill group, they ranged from 0.09 ± 0.15 mm to 0.83 ± 0.23 mm. At the apical tip, the overall deviations in the print group ranged from 0.14 ± 0.56 mm to 1.27 ± 0.66 mm, whereas in the mill group, they ranged from 0.15 ± 0.57 mm to 1.09 ± 0.44 mm. The mean intra-class and inter-class correlation coefficients ranged from 0.904 to 0.987. No statistically significant differences were found between the groups. ConclusionsCAD/CAM milled guides yielded spatial and angular accuracies comparable to those of 3D printed guides with notable deviations in both in the vertical positioning of TADs. Clinical SignificanceDigital planning of orthodontic temporary implants combines clinical predictability and the safety of surrounding tissue. Therefore, the transfer accuracy of the guides is crucial. This preclinical study was the first to evaluate CAD/CAM milling for orthodontic guides and found its accuracy comparable to that of the current "gold standard".

Validation of Polar ElixirTM Pulse Oximeter against Arterial Blood Gases during Stepwise Steady State Inspired Hypoxia.

The purpose of this study was to evaluate the accuracy of peripheral oxygen saturation (SpO2) measurements from Polar ElixirTM pulse oximetry technology compared to arterial oxygen saturation (SaO2) measurements during acute stepwise steady state inspired hypoxia at rest. A post hoc objective was to determine if SpO2 measurements could be improved by recalibrating the Polar ElixirTM algorithm with SaO2 values from a random subset of participants. The International Organization for Standardization (ISO) protocol (ISO 80601-2-61:2017) for evaluating the SpO2 accuracy of pulse oximeter equipment was followed whereby five plateaus of SaO2 between 70-100% were achieved using stepwise reductions in inspired O2 during supine rest. Blood samples drawn through a radial arterial catheter from 25 participants were first used to compare SaO2 to SpO2 measurements from Polar ElixirTM. Then the Polar ElixirTM algorithm was recalibrated using SaO2 data from 13 random participants and SpO2 estimates were recalculated for the other 12 participants. For SaO2 values between 70-100%, root mean square error (RMSE), intraclass correlations (ICC), Pearson correlations, and Bland-Altman plots were used to assess the accuracy, agreement, and strength of relationship between SaO2 values and SpO2 values from Polar ElixirTM. The initial RMSE for Polar ElixirTM was 4.13%. After recalibrating the algorithm, the RMSE was improved to 2.67%. The ICC revealed excellent levels of agreement between SaO2 and Polar ElixirTM SpO2 values both before (ICC(3,1) = 0.837, df = 574, p < 0.001) and after (ICC(3,1) = 0.942, df = 287, p < 0.001) recalibration. Relative to ISO standards, Polar ElixirTM yielded accurate SpO2 measurements during stepwise inspired hypoxia at rest when compared to SaO2 values, which were improved by recalibrating the algorithm using a subset of the SaO2 data.

SAEM systematic online academic resource (SOAR) review: Gastrointestinal illnesses.

Free open access medical education (FOAM) has become an essential tool for emergency medicine (EM) education and can be valuable to clinicians as a point-of-care resource. The development of the revised Medical Education Translational Resources Impact and Quality (rMETRIQ) tool provides a standardized means of quality assessment. Previous entries of the Society for Academic Emergency Medicine systematic online academic resource (SOAR) series have focused on renal, endocrine, and sickle cell disorders. In this iteration, we strive to identify, curate, and describe FOAM topics specific to acute gastrointestinal (GI) illnesses. We searched 389 keywords across 11 GI topics that were modified from the 2019 Model of the Clinical Practice of EM (EM Model) using the search engine Google FOAM and within the top 50 websites listed on Academic Life in Emergency Medicine's Social Media Index. The sites underwent preliminary screening to eliminate resources that were not relevant to EM or GI illnesses. Identified resources were evaluated with the rMETRIQ tool by five board-certified EM physicians who received rMETRIQ tool rater training. After duplicates of the initial 39,505 resources were eliminated, 8059 remained. Primary screening resulted in a final 1202 resources. The most common categories were large bowel (18%), small bowel (13%), stomach (11%), esophagus (11%), biliary (11%), and liver (10%). Many resources covered multiple topics and subtopics. The final mean intraclass correlation coefficient among the five physicians was 0.95 (95% CI 0.92-0.98) for rMETRIQ scoring. We identified 256 sites considered "high quality" with a rMETRIQ score of 16 or higher as designated in prior reviews. This iteration of the SOAR review resulted in the highest number of high-quality resources compared to other SOAR reviews, with 21% of resources thus far scoring ≥ 16. A final list of high-quality resources can guide trainees, educator recommendations, and FOAM authors.

Deep learning-based multimodal segmentation of oropharyngeal squamous cell carcinoma on CT and MRI using self-configuring nnU-Net.

To evaluate deep learning-based segmentation models for oropharyngeal squamous cell carcinoma (OPSCC) using CT and MRI with nnU-Net. This single-center retrospective study included 91 patients with OPSCC. The patients were grouped into the development (n = 56), test 1 (n = 13), and test 2 (n = 22) cohorts. In the development cohort, OPSCC was manually segmented on CT, MR, and co-registered CT-MR images, which served as the ground truth. The multimodal and multichannel input images were then trained using a self-configuring nnU-Net. For evaluation metrics, dice similarity coefficient (DSC) and mean Hausdorff distance (HD) were calculated for test cohorts. Pearson's correlation and Bland-Altman analyses were performed between ground truth and prediction volumes. Intraclass correlation coefficients (ICCs) of radiomic features were calculated for reproducibility assessment. All models achieved robust segmentation performances with DSC of 0.64 ± 0.33 (CT), 0.67 ± 0.27 (MR), and 0.65 ± 0.29 (CT-MR) in test cohort 1 and 0.57 ± 0.31 (CT), 0.77 ± 0.08 (MR), and 0.73 ± 0.18 (CT-MR) in test cohort 2. No significant differences were found in DSC among the models. HD of CT-MR (1.57 ± 1.06mm) and MR models (1.36 ± 0.61mm) were significantly lower than that of the CT model (3.48 ± 5.0mm) (p = 0.037 and p = 0.014, respectively). The correlation coefficients between the ground truth and prediction volumes for CT, MR, and CT-MR models were 0.88, 0.93, and 0.9, respectively. MR models demonstrated excellent mean ICCs of radiomic features (0.91-0.93). The self-configuring nnU-Net demonstrated reliable and accurate segmentation of OPSCC on CT and MRI. The multimodal CT-MR model showed promising results for the simultaneous segmentation on CT and MRI. Deep learning-based automatic detection and segmentation of oropharyngeal squamous cell carcinoma on pre-treatment CT and MRI would facilitate radiologic response assessment and radiotherapy planning. • The nnU-Net framework produced a reliable and accurate segmentation of OPSCC on CT and MRI. • MR and CT-MR models showed higher DSC and lower Hausdorff distance than the CT model. • Correlation coefficients between the ground truth and predicted segmentation volumes were high in all the three models.

Automated MRI liver segmentation for anatomical segmentation, liver volumetry, and the extraction of radiomics.

To develop and evaluate a deep convolutional neural network (DCNN) for automated liver segmentation, volumetry, and radiomic feature extraction on contrast-enhanced portal venous phase magnetic resonance imaging (MRI). This retrospective study included hepatocellular carcinoma patients from an institutional database with portal venous MRI. After manual segmentation, the data was randomly split into independent training, validation, and internal testing sets. From a collaborating institution, de-identified scans were used for external testing. The public LiverHccSeg dataset was used for further external validation. A 3D DCNN was trained to automatically segment the liver. Segmentation accuracy was quantified by theDice similarity coefficient (DSC) with respect to manual segmentation. A Mann-Whitney U test was used to compare the internal and external test sets. Agreement of volumetry and radiomic features was assessed using the intraclass correlation coefficient (ICC). In total, 470 patients met the inclusion criteria (63.9±8.2 years; 376 males) and 20 patients were used for external validation (41±12 years; 13 males). DSC segmentation accuracy of the DCNN was similarly high between the internal (0.97±0.01) and external (0.96±0.03) test sets (p=0.28) and demonstrated robust segmentation performance on public testing (0.93±0.03). Agreement of liver volumetry was satisfactory in the internal (ICC, 0.99), external (ICC, 0.97), and public (ICC, 0.85) test sets. Radiomic features demonstrated excellent agreement in the internal (mean ICC, 0.98±0.04), external (mean ICC, 0.94±0.10), and public (mean ICC, 0.91±0.09) datasets. Automated liver segmentation yields robust and generalizable segmentation performance on MRI data and can be used for volumetry and radiomic feature extraction. Liver volumetry, anatomic localization, and extraction of quantitative imaging biomarkers require accurate segmentation, but manual segmentation is time-consuming. A deep convolutional neural network demonstrates fast and accurate segmentation performance on T1-weighted portal venous MRI. • This deep convolutional neural network yields robust and generalizable liver segmentation performance on internal, external, and public testing data. • Automated liver volumetry demonstrated excellent agreement with manual volumetry. • Automated liver segmentations can be used for robust and reproducible radiomic feature extraction.

The hemodynamic cardiac profiler volume-time curves and related parameters: an MRI validation study

Background. The hemodynamic cardiac profiler (HCP) is a new, non-invasive, operator-independent screening tool that uses six independent electrode pairs on the frontal thoracic skin, and a low-intensity, patient-safe, high-frequency applied alternating current to measure ventricular volume dynamics during the cardiac cycle for producing ventricular volume-time curves (VTCs). Objective. To validate VTCs from HCP against VTCs from MRI in healthy volunteers. Approach. Left- and right-ventricular VTCs were obtained by HCP and MRI in six healthy participants in supine position. Since HCP is not compatible with MRI, HCP measurements were performed within 20 min before and immediately after MRI, without intermittent fluid intake or release by participants. Intraclass correlation coefficients (ICCs) were calculated to validate HCP-VTC against MRI-VTC and to assess repeatability of HCP measurements before and after MRI. Bland–Altman plots were used to assess agreement between relevant HCP- and MRI-VTC-derived parameters. Precision of HCP’s measurement of VTC-derived parameters was determined for each study participant by calculating the coefficients of variation and repeatability coefficients. Main results. Left- and right-ventricular VTC ICCs between HCP and MRI were >0.8 for all study participants, indicating excellent agreement between HCP-VTCs and MRI-VTCs. Mean (range) ICC of HCP right-ventricular VTC versus MRI right-ventricular VTC was 0.94 (0.88–0.99) and seemed to be slightly higher than the mean ICC of HCP left-ventricular VTC versus MRI-VTC (0.91 (0.80–0.96)). The repeatability coefficient for HCP’s measurement of systolic time (tSys) was 45.0 ms at a mean value of 282.9 ± 26.3 ms. Repeatability of biventricular HCP-VTCs was excellent (ICC 0.96 (0.907–0.995)). Significance. Ventricular volume dynamics measured by HCP-VTCs show excellent agreement with VTCs measured by MRI. Since abnormal tSys is a sign of numerous cardiac diseases, the HCP may potentially be used as a diagnostic screening tool.

Inter-observer variance of examiner scoring in urology Objective Structured Clinical Examinations.

The Objective Structured Clinical Examination (OSCE) is an attractive tool of competency assessment in a high-stakes summative exam. An advantage of the OSCE is the ability to assess more realistic context, content, and procedures. Each year, the Queen's Urology Exam Skills Training (QUEST) is attended by graduating Canadian urology residents to simulate their upcoming board exams. The exam consists of a written component and an OSCE. The aim of this study was to determine the inter-observer consistency of scoring between two examiners of an OSCE for a given candidate. Thirty-nine participants in 2020 and 37 participants in 2021 completed four stations of OSCEs virtually over the Zoom platform. Each candidate was examined and scored independently by two different faculty urologists in a blinded fashion at each station. The OSCE scoring consisted of a checklist rating scale for each question. An intra-class correlation (ICC) analysis was conducted to determine the inter-rater reliability of the two examiners for each of the four OSCE stations in both the 2020 and 2021 OSCEs. For the 2020 data, the prostate cancer station scores were most strongly correlated (ICC 0.746, 95% confidence interval [CI] 0.556-0.862, p<0.001). This was followed by the general urology station (ICC 0.688, 95% CI 0.464-0.829, p<0.001), the urinary incontinence station (ICC 0.638, 95% CI 0.403-0.794, p<0.001), and finally the nephrolithiasis station (ICC 0.472, 95% CI 0.183-0.686, p<0.001). For the 2021 data, the renal cancer station had the highest ICC at 0.866 (95% CI 0.754-0.930, p<0.001). This was followed by the nephrolithiasis station (ICC 0.817, 95% CI 0.673-0.901, p<0.001), the pediatric station (ICC 0.809, 95% CI 0.660-0.897, p<0.001), and finally the andrology station (ICC 0.804, 95% CI 0.649-0.895, p<0.001). A Pearson correlation coefficient was calculated for all stations, and all show a positive correlation with global exam scores. It is noteworthy that some stations were more predictive of overall performance, but this did not necessarily mean better ICC scores for these stations. Given a specific clinical scenario in an OSCE exam, inter-rater reliability of scoring can be compromised on occasion. Care should be taken when high-stakes decisions about promotion are made based on OSCEs with limited standardization.

Computer-assisted assessment of segmental bimaxillary surgery using voxel- and surface-based registration: A comparative study

The purpose of the present study was to compare the precision and reliability of voxel- and surface-based registration for computer-assisted assessment of the surgical accuracy and postoperative stability of segmental bimaxillary surgery. Three-dimensional translational and rotational measurements were performed by two observers using voxel- and surface-based registration. The precision and reliability of the measurements were calculated by the mean absolute differences (MAD) and intraclass correlation coefficients (ICC) at 95 % confidence intervals. A paired t-test or the non-parametric equivalent, Wilcoxon signed-rank test, was applied to statistically evaluate whether the precision of voxel- and surface-based registration was statistically significantly different (p < 0.05). Voxel-based registration had high precision (MAD <0.44 mm/0.92°) and excellent reliability, ICC [0.82–1.00]. The precision of surface-based registration was lower (MAD <0.56 mm/1.45°) and the reliability ranged from poor to excellent for the different bone segments, ICC [0.33–1.00]. Both registration techniques had high precision and excellent reliability for the assessment of the surgical accuracy, and the error margin of both techniques was clinical irrelevant. However, the increased precision of voxel-based registration was statistically significant (p < 0.05) for the maxillary segments and the chin, and the stability measurement error (ranging up to 1.58 mm and 4.46°) introduced by surface-based registration may be considered clinical relevant for these bone segments. Within the limitations of the present comparative study, voxel-based registration generally exhibited higher precision and reliability than surface-based registration for the surgical accuracy and postoperative stability assessment of segmental bimaxillary surgery.

Comparison of deep learning-based image segmentation methods for intravascular ultrasound on retrospective and large image cohort study

ObjectivesThe aim of this study was to investigate the generalization performance of deep learning segmentation models on a large cohort intravascular ultrasound (IVUS) image dataset over the lumen and external elastic membrane (EEM), and to assess the consistency and accuracy of automated IVUS quantitative measurement parameters.MethodsA total of 11,070 IVUS images from 113 patients and pullbacks were collected and annotated by cardiologists to train and test deep learning segmentation models. A comparison of five state of the art medical image segmentation models was performed by evaluating the segmentation of the lumen and EEM. Dice similarity coefficient (DSC), intersection over union (IoU) and Hausdorff distance (HD) were calculated for the overall and for subsets of different IVUS image categories. Further, the agreement between the IVUS quantitative measurement parameters calculated by automatic segmentation and those calculated by manual segmentation was evaluated. Finally, the segmentation performance of our model was also compared with previous studies.ResultsCENet achieved the best performance in DSC (0.958 for lumen, 0.921 for EEM) and IoU (0.975 for lumen, 0.951 for EEM) among all models, while Res-UNet was the best performer in HD (0.219 for lumen, 0.178 for EEM). The mean intraclass correlation coefficient (ICC) and Bland–Altman plot demonstrated the extremely strong agreement (0.855, 95% CI 0.822–0.887) between model's automatic prediction and manual measurements.ConclusionsDeep learning models based on large cohort image datasets were capable of achieving state of the art (SOTA) results in lumen and EEM segmentation. It can be used for IVUS clinical evaluation and achieve excellent agreement with clinicians on quantitative parameter measurements.

Assessment of landmark detection in cephalometric radiographs with different conditions of brightness and contrast using the an artificial intelligence software.

To evaluate the reliability and reproducibility of an artificial intelligence (AI) software in identifying cephalometric points on lateral cephalometric radiographs considering four settings of brightness and contrast. Brightness and contrast of 30 lateral cephalometric radiographs were adjusted into four different settings. Then, the control examiner (ECont), the calibrated examiner (ECal), and the CEFBOT AI software (AIs) each marked 19 cephalometric points on all radiographs. Reliability was assessed with a second analysis of the radiographs 15 days after the first one. Statistical significance was set at p < 0.05. Reliability of landmark identification was excellent for the human examiners and the AIs regardless of the type of brightness and contrast setting (mean intraclass correlation coefficient >0.89). When ECont and ECal were compared for reproducibility, there were more cephalometric points with significant differences on the x-axis of the image with the highest contrast and the lowest brightness, namely N(p = 0.033), S(p = 0.030), Po(p < 0.001), and Pog'(p = 0.012). Between ECont and AIs, there were more cephalometric points with significant differences on the image with the highest contrast and the lowest brightness, namely N(p = 0.034), Or(p = 0.048), Po(p < 0.001), A(p = 0.042), Pog'(p = 0.004), Ll(p = 0.005), Ul(p < 0.001), and Sn(p = 0.001). While the reliability of the AIs for cephalometric landmark identification was rated as excellent, low brightness and high contrast seemed to affect its reproducibility. The experienced human examiner, on the other hand, did not show such faulty reproducibility; therefore, the AIs used in this study is an excellent auxiliary tool for cephalometric analysis, but still depends on human supervision to be clinically reliable.

Test-Retest Reliability of a Physical Activity Behavior, Health and Wellbeing Questionnaire in Adolescents

Background: The aim of this study was to examine the test-retest reliability of the physical activity behavior, health and wellbeing questionnaire, in adolescent populations, administered by teachers in school settings, in the Republic of Ireland.   Methods: A cross-sectional, mixed sample of 55 participants (45.5% males: Age, 13.94 (±.40) were included. The participants completed the questionnaire on two occasions (T1 and T2), on the same day and time, one week apart following identical procedures. Variables for testing included physical activity behavior (n=13), health (n=11) and wellbeing (n=2). Test-retest reliability of the questionnaire’s covariates, including family affluence and physical impairments were also examined. Results: Systematic error (Bland-Altman plots) was found to be near to zero for each of the physical activity behavior, health and wellbeing variables. The combined mean coefficient of variation was lower for females (10.19%) in comparison to males (13.01%). Similarly, the combined mean intraclass correlation coefficients were higher for females (&gt;.901) than males (&gt;.822).   Conclusions: This study found the physical activity behavior, health and wellbeing questionnaire to be reliable for use in adolescent populations.

A validation study for wide-range remote assessment of cognitive functions in the healthy older Japanese population: a pilot randomised crossover trial

BackgroundThe assessment of a wide range of cognitive functions using video teleconference (VTC) systems cannot be applied in practice yet. We aimed to determine the feasibility and reliability of previously unvalidated remote cognitive function tests in Japan using common information and communication technology (ICT) devices, software, and VTC systems compared with face-to-face (FTF) assessment.MethodsThe sample consisted of 26 participants from senior citizens clubs and an employment service centre in Sapporo Japan, including 11 females and 15 males (age averaged 78.6 ± 6.8 years). Tests included the RCPM, Story recall, 10/36 spatial recall, selective reminding test, SDMT, PASAT, FAB, TMT-A, TMT-B, visual cancellation task, digit span, tapping span. The experimental design was a counterbalanced crossover randomised controlled trial. Intraclass correlations (ICCs), paired-samples t-tests, Cohen’s Kappa (κ) coefficients, and Wilcoxon signed-rank test were calculated to compare the scores between VTC and FTF assessments.ResultsAll ICCs were significant and ranged from 0.47 (RCPM time) to 0.92 (RCPM score and PASAT), with a mean ICC of 0.75. Digit span using Cohen’s Kappa (κ) coefficient was significant, but the tapping span was not. Paired samples t-test showed statistically significant differences in SDMT, RCPM time, and cancellation time.ConclusionsThe results suggest that remote video conference-based neuropsychological tests even using familiar devices and software may be able to assess a wide range of cognitive functions in the Japanese older population. As for the processing speed tasks, we need to create our own standards for the remote condition. For the tapping span, we should consider increasing the number of trials.

Test-Retest Reliability of Functional Connectivity in Adolescents With Depression

BackgroundThe test-retest reliability of functional magnetic resonance imaging is critical to identifying reproducible biomarkers for psychiatric illness. Recent work has shown how reliability limits the observable effect size of brain-behavior associations, hindering detection of these effects. However, while a fast-growing literature has explored both univariate and multivariate reliability in healthy individuals, relatively few studies have explored reliability in populations with psychiatric illnesses or how this interacts with age. MethodsHere, we investigated functional connectivity reliability over the course of 1 year in a longitudinal cohort of 88 adolescents (age at baseline = 15.63 ± 1.29 years; 64 female) with major depressive disorder (MDD) and without MDD (healthy volunteers [HVs]). We compared a univariate metric, intraclass correlation coefficient, and 2 multivariate metrics, fingerprinting and discriminability. ResultsAdolescents with MDD had marginally higher mean intraclass correlation coefficient (μMDD = 0.34, 95% CI, 0.12–0.54; μHV = 0.27, 95% CI, 0.05–0.52), but both groups had poor average intraclass correlation coefficients (<0.4). Fingerprinting index was greater than chance and did not differ between groups (fingerprinting indexMDD = 0.75; fingerprinting indexHV = 0.91; Poisson tests p < .001). Discriminability indicated high multivariate reliability in both groups (discriminabilityMDD = 0.80; discriminabilityHV = 0.82; permutation tests p < .01). Neither univariate nor multivariate reliability was associated with symptom severity or edge-level effect size of group differences. ConclusionsOverall, we found little evidence for a relationship between depression and reliability of functional connectivity during adolescence. These findings suggest that biomarker identification in depression is not limited due to reliability compared with healthy samples and support the shift toward multivariate analysis for improved power and reliability.

Validation of Wearable Digital Devices for Heart Rate Measurement During Exercise Test in Patients With Coronary Artery Disease.

To assess the accuracy of recently commercialized wearable devices in heart rate (HR) measurement during cardiopulmonary exercise test (CPX) under gradual increase in exercise intensity, while wearable devices with HR monitors are reported to be less accurate in different exercise intensities. CPX was performed for patients with coronary artery disease (CAD). Twelve lead electrocardiograph (ECG) was the gold standard and Apple watch 7 (AW7), Galaxy watch 4 (GW4) and Bio Patch Mobicare 200 (MC200) were applied for comparison. Paired absolute difference (PAD), mean absolute percentage error (MAPE) and intraclass correlation coefficient (ICC) were evaluated for each device. Forty-four participants with CAD were included. All the devices showed MAPE under 2% and ICC above 0.9 in rest, exercise and recovery phases (MC200=0.999, GW4=0.997, AW7=0.998). When comparing exercise and recovery phase, PAD of MC200 and AW7 in recovery phase were significantly bigger than PAD of exercise phase (p<0.05). Although not significant, PAD of GW4 tended to be bigger in recovery phase, too. Also, when stratified by HR 20, ICC of all the devices were highest under HR of 100, and ICC decreased as HR increased. However, except for ICC of GW4 at HR above 160 (=0.867), all ICCs exceeded 0.9 indicating excellent accuracy. The HR measurement of the devices validated in this study shows a high concordance with the ECG device, so CAD patients may benefit from the devices during high-intensity exercise under conditions where HR is measured reliably.

The velocity of resistance exercise does not accurately assess repetitions-in-reserve

ABSTRACT This study assessed the reliability of mean concentric bar velocity from 3- to 0-repetitions in reserve (RIR) across four sets in different exercises (bench press and prone row) and with different loads (60 and 80% 1-repetition maximum; 1RM). Whether velocity values from set one could be used to predict RIR in subsequent sets was also examined. Twenty recreationally active males performed baseline 1RM testing before two randomised sessions of four sets to failure with 60 or 80% 1RM. A linear position transducer measured mean concentric velocity of repetitions, and the velocity associated with each RIR value up to 0-RIR. For both exercises, velocity decreased between each repetition from 3- to 0-RIR (p ≤ 0.010). Mean concentric velocity of RIR values was not reliable across sets in the bench press (mean intraclass correlation coefficient [ICC] = 0.40, mean coefficient of variation [CV] = 21.3%), despite no significant between-set differences (p = 0.530). Better reliability was noted in the prone row (mean ICC = 0.80, mean CV = 6.1%), but velocity declined by 0.019–0.027 m·s−1 (p = 0.032) between sets. Mean concentric velocity was 0.050–0.058 m·s−1 faster in both exercises with 60% than 80% 1RM with (p < 0.001). At the individual level, the velocity of specific RIR values from set one accurately predicted RIR from 5- to 0-RIR for 30.9% of repetitions in subsequent sets. These findings suggest that velocity of specific RIR values vary across exercises, loads and sets. As velocity-based RIR estimates were not accurate for 69.1% of repetitions, alternative methods to should be considered for autoregulating of resistance exercise in recreationally active individuals.

Artificial Intelligence‐Powered Acoustic Analysis System for Dysarthria Severity Assessment

Dysarthria is common in movement disorders, such as Wilson's disease (WD), Parkinson's disease, or Huntington's disease. Dysarthria severity assessment is often indispensable for the management of these diseases. However, such assessment is usually labor‐intensive, time‐consuming, and expensive. To seek efficient and cost‐effective solutions for dysarthria assessment, an artificial intelligence (AI)‐powered acoustic analysis system is proposed and its performance in a valuable sample of WD, an ideal disease model with mainly mixed dysarthria, is verified. A test‐retest reliability analysis yields excellent reproducibility in the acoustic measures (mean intraclass correlation coefficient [ICC] = 0.81). Then, a system for dysarthria assessment is trained with WD patients (n = 65) and sex‐matched healthy controls (n = 65) using a machine learning approach. The system achieves reasonable performance in evaluating dysarthria severity with either stepwise classification or regression (all areas under the curve &gt;80%; mean absolute error = 6.25, r = 0.79, p &lt; 0.0001). The diadochokinesis and sustained phonation tasks contribute the most to prediction, and the corresponding acoustic features can provide significant and independent contributions. The present study demonstrates the feasibility and good performance of the AI‐powered acoustic analysis framework, offering the potential to facilitate early screening and subsequent management of dysarthria.

Precision and Accuracy Assessment of Cephalometric Analyses Performed by Deep Learning Artificial Intelligence with and without Human Augmentation

The aim was to assess the precision and accuracy of cephalometric analyses performed by artificial intelligence (AI) with and without human augmentation. Four dental professionals with varying experience levels identified 31 landmarks on 30 cephalometric radiographs twice. These landmarks were re-identified by all examiners with the aid of AI. Precision and accuracy were assessed by using intraclass correlation coefficients (ICCs) and mean absolute errors (MAEs). AI revealed the highest precision, with a mean ICC of 0.97, while the dental student had the lowest (mean ICC: 0.77). The AI/human augmentation method significantly improved the precision of the orthodontist, resident, dentist, and dental student by 3.26%, 2.17%, 19.75%, and 23.38%, respectively. The orthodontist demonstrated the highest accuracy with an MAE of 1.57 mm/°. The AI/human augmentation method improved the accuracy of the orthodontist, resident, dentist, and dental student by 12.74%, 19.10%, 35.69%, and 33.96%, respectively. AI demonstrated excellent precision and good accuracy in automated cephalometric analysis. The precision and accuracy of the examiners with the aid of AI improved by 10.47% and 27.27%, respectively. The AI/human augmentation method significantly improved the precision and accuracy of less experienced dental professionals to the level of an experienced orthodontist.

Modality matters for health disparity groups on the Montreal Cognitive Assessment (MoCA): Findings from the Einstein Aging Study (EAS)

AbstractBackgroundThe Telephone‐Montreal Cognitive Assessment (T‐MoCA) is a widely‐used, remotely administered cognitive screen. The T‐MoCA demonstrates adequate psychometric properties and may be used when standard, in‐person administration of the MoCA is not feasible (Katz et al., 2020). There is a critical need to investigate the agreement of the T‐MoCA across administration modalities (in‐person versus telehealth) among ethnically and racially diverse populations.MethodsThe Einstein Aging Study (EAS) includes a community‐based of racially/ethnically diverse community‐dwelling individuals, ≥ age 70 from the Bronx, NY. Participants (N = 424) were free of dementia at enrollment and completed the T‐MoCA along with in‐person neuropsychological tests including the MoCA‐22, which excludes items unsuitable for telephone administration paralleling the T‐MoCA (Table 1). Intraclass correlation coefficients (ICCs) between T‐MoCA and MoCA‐22 were used to measure agreement. Linear mixed effects models were applied to compare mean scores and ICCs among ethnic/racial groups, with and without adjusting for age, gender, education and depressive symptoms.ResultsParticipants were on average 78.2 years, 65.8% female, 47.6% non‐Hispanic White, 38.0% non‐Hispanic Black, and 14.4% Hispanic. There was no significant difference between mean T‐MoCA and MoCA‐22 scores and no differential associations with age, gender, education, and depressive symptoms. For both modalities, NH Black (difference ‐1.54, p&lt;.0001) and Hispanic (difference ‐2.04, p&lt;.0001) groups demonstrated significantly lower scores compared to non‐Hispanic Whites (Table 2). As shown in Table 2, ICCs among non‐Hispanic White, non‐Hispanic Black, and Hispanic groups were 0.64, 0.50 and 0.42, respectively, and comparisons of ICCs to non‐Hispanic White were significantly lower in non‐Hispanic Black (p = 0.0495) and borderline significant in Hispanic participants (p = 0.0552). Findings were similar after adjusting for covariates.ConclusionThe agreement between telephone and in‐person modalities of the MoCA was weaker in non‐Hispanic Black and Hispanic participants (compared to non‐Hispanic White). Further investigation of factors and/or items that contribute to these lower correlations is warranted. Although the T‐MoCA and other remotely administered cognitive screens have the potential to broaden accessibility to marginalized groups, researchers must use caution when developing or selecting appropriate demographically adjusted norms and interpreting scores of diverse older adults.