Inter-expert and intra-expert reliability in sleep spindle scoring.
Ontology highlight
ABSTRACT:
SUBMITTER: Wendt SL
PROVIDER: S-EPMC4426257 | biostudies-literature | 2015 Aug
REPOSITORIES: biostudies-literature
Publications
Inter-expert and intra-expert reliability in sleep spindle scoring.
Wendt Sabrina L SL Welinder Peter P Sorensen Helge B D HB Peppard Paul E PE Jennum Poul P Perona Pietro P Mignot Emmanuel E Warby Simon C SC
Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology 20141110 8
<h4>Objectives</h4>To measure the inter-expert and intra-expert agreement in sleep spindle scoring, and to quantify how many experts are needed to build a reliable dataset of sleep spindle scorings.<h4>Methods</h4>The EEG dataset was comprised of 400 randomly selected 115s segments of stage 2 sleep from 110 sleeping subjects in the general population (57±8, range: 42-72 years). To assess expert agreement, a total of 24 Registered Polysomnographic Technologists (RPSGTs) scored spindles in a subse ...[more]
Similar Datasets
Project description:BackgroundSleep spindles are a marker of stage 2 NREM sleep that are linked to learning & memory and are altered by many neurological diseases. Although visual inspection of the EEG is considered the gold standard for spindle detection, it is time-consuming, costly and can introduce inter/ra-scorer bias.New methodOur goal was to develop a simple and efficient sleep-spindle detector (algorithm #7, or 'A7') that emulates human scoring. 'A7' runs on a single EEG channel and relies on four parameters: the absolute sigma power, relative sigma power, and correlation/covariance of the sigma band-passed signal to the original EEG signal. To test the performance of the detector, we compared it against a gold standard spindle dataset derived from the consensus of a group of human experts.ResultsThe by-event performance of the 'A7' spindle detector was 74% precision, 68% recall (sensitivity), and an F1-score of 0.70. This performance was equivalent to an individual human expert (average F1-score = 0.67).Comparison with existing method(s)The F1-score of 'A7' was 0.17 points higher than other spindle detectors tested. Existing detectors have a tendency to find large numbers of false positives compared to human scorers. On a by-subject basis, the spindle density estimates produced by A7 were well correlated with human experts (r2 = 0.82) compared to the existing detectors (average r2 = 0.27).ConclusionsThe 'A7' detector is a sensitive and precise tool designed to emulate human spindle scoring by minimizing the number of 'hidden spindles' detected. We provide an open-source implementation of this detector for further use and testing.
| S-EPMC6415669 | biostudies-literature
Project description:ObjectivesThe radiological Pettersson score (PS) is widely applied for classification of arthropathy to evaluate costly haemophilia treatment. This study aims to assess and improve inter- and intra-observer reliability and agreement of the PS.MethodsTwo series of X-rays (bilateral elbows, knees, and ankles) of 10 haemophilia patients (120 joints) with haemophilic arthropathy were scored by three observers according to the PS (maximum score 13/joint). Subsequently, (dis-)agreement in scoring was discussed until consensus. Example images were collected in an atlas. Thereafter, second series of 120 joints were scored using the atlas. One observer rescored the second series after three months. Reliability was assessed by intraclass correlation coefficients (ICC), agreement by limits of agreement (LoA).ResultsMedian Pettersson score at joint level (PSjoint) of affected joints was 6 (interquartile range 3-9). Using the consensus atlas, inter-observer reliability of the PSjoint improved significantly from 0.94 (95 % confidence interval (CI) 0.91-0.96) to 0.97 (CI 0.96-0.98). LoA improved from ±1.7 to ±1.1 for the PSjoint. Therefore, true differences in arthropathy were differences in the PSjoint of >2 points. Intra-observer reliability of the PSjoint was 0.98 (CI 0.97-0.98), intra-observer LoA were ±0.9 points.ConclusionsReliability and agreement of the PS improved by using a consensus atlas.Key points• Reliability of the Pettersson score significantly improved using the consensus atlas. • The presented consensus atlas improved the agreement among observers. • The consensus atlas could be recommended to obtain a reproducible Pettersson score.
| S-EPMC4869743 | biostudies-literature
Project description:Objectives:Scoring laboratory polysomnography (PSG) data remains a manual task of visually annotating 3 primary categories: sleep stages, sleep disordered breathing, and limb movements. Attempts to automate this process have been hampered by the complexity of PSG signals and physiological heterogeneity between patients. Deep neural networks, which have recently achieved expert-level performance for other complex medical tasks, are ideally suited to PSG scoring, given sufficient training data. Methods:We used a combination of deep recurrent and convolutional neural networks (RCNN) for supervised learning of clinical labels designating sleep stages, sleep apnea events, and limb movements. The data for testing and training were derived from 10 000 clinical PSGs and 5804 research PSGs. Results:When trained on the clinical dataset, the RCNN reproduces PSG diagnostic scoring for sleep staging, sleep apnea, and limb movements with accuracies of 87.6%, 88.2% and 84.7% on held-out test data, a level of performance comparable to human experts. The RCNN model performs equally well when tested on the independent research PSG database. Only small reductions in accuracy were noted when training on limited channels to mimic at-home monitoring devices: frontal leads only for sleep staging, and thoracic belt signals only for the apnea-hypopnea index. Conclusions:By creating accurate deep learning models for sleep scoring, our work opens the path toward broader and more timely access to sleep diagnostics. Accurate scoring automation can improve the utility and efficiency of in-lab and at-home approaches to sleep diagnostics, potentially extending the reach of sleep expertise beyond specialty clinics.
| S-EPMC6289549 | biostudies-literature
Project description:The Posturomed device is used as a scientific tool to quantify human dynamic balance ability due to unexpected perturbations, and as a training device. Consequently, the question arises whether such measurements are compromised by learning effects. Therefore, this study aimed to analyze inter- and intra-day reliability of dynamic balance responses using the Posturomed. Thirty healthy young subjects participated (24.3±3.2 years). The Posturomed was equipped with a triggering mechanism to enable unexpected, horizontal platform perturbations. A force platform was used to quantify Center of Pressure (COP) excursions for two time intervals: interval 1 (0-70 ms post perturbation) and interval 2 (71-260 ms post perturbation). Dynamic balance tests were performed in single leg stances in medio-lateral and anterior-posterior perturbation directions. Inter- and intra-day reliability were assessed descriptively using Bland-Altman plots and inferentially using tests for systematic error and intra-class-correlations. With regard to the mean COP excursions for every subject and all intervals, some cases revealed significant differences between measurement sessions, however, none were considered relevant. Furthermore, intra class correlation coefficients reflected high magnitudes, which leads to the assumption of good relative reliability. However, analyzing inter- and intra-day reliability using Bland-Altman plots revealed one exception: intra-day comparisons for the anterior-posterior direction in interval 2, which points towards possible learning effects. In summary, results reflected good overall reliability with the exception of certain intra-day comparisons in the anterior-posterior perturbation direction, which could indicate learning effects in those particular conditions.
| S-EPMC4560445 | biostudies-literature
Project description:Foot pain drawings (manikins) are commonly used to describe foot pain location in self-report health surveys. Respondents shade the manikin where they experience pain. The manikin is then scored via a transparent overlay that divides the drawings into areas. In large population based studies they are often scored by multiple raters. A difference in how different raters score manikins (inter-rater repeatability), or in how an individual rater scores manikins over time (intra-rater repeatability) can therefore affect data quality. This study aimed to assess inter- and intra-rater repeatability of scoring of the foot manikin.A random sample was generated of 50 respondents to a large population based survey of adults aged 50 years and older who experienced foot pain and completed a foot manikin. Manikins were initially scored by any one of six administrative staff (Rating 1). These manikins were re-scored by a second rater (Rating 2). The second rater then re-scored the manikins one week later (Rating 3). The following scores were compared: Rating 1 versus Rating 2 (inter-rater repeatability), and Rating 2 versus Rating 3 (intra-rater repeatability). A novel set of clinically relevant foot pain regions made up of one or more individual areas on the foot manikin were developed, and assessed for inter- and intra-rater repeatability.Scoring agreement of 100% (all 50 manikins) was seen in 69% (40 out of 58) of individual areas for inter-rater scoring (range 94 to 100%), and 81% (47 out of 58) of areas for intra-rater scoring (range 96 to 100%). All areas had a kappa value of ?0.70 for inter- and intra-rater scoring. Scoring agreement of 100% was seen in 50% (10 out of 20) of pain regions for inter-rater scoring (range 96 to 100%), and 95% (19 out of 20) of regions for intra-rater scoring (range 98 to 100%). All regions had a kappa value of >0.70 for inter- and intra-rater scoring.Individual and multiple raters can reliably score the foot pain manikin. In addition, our proposed regions may be used to reliably classify different patterns of foot pain using the foot manikin.
| S-EPMC3831824 | biostudies-other
Project description:PurposeFirst, to evaluate inter-rater reliability when human raters estimate the reading performance of visually impaired individuals using the MNREAD acuity chart. Second, to evaluate the agreement between computer-based scoring algorithms and compare them with human rating.MethodsReading performance was measured for 101 individuals with low vision, using the Portuguese version of the MNREAD test. Seven raters estimated the maximum reading speed (MRS) and critical print size (CPS) of each individual MNREAD curve. MRS and CPS were also calculated automatically for each curve using two different algorithms: the original standard deviation method (SDev) and a non-linear mixed effects (NLME) modeling. Intra-class correlation coefficients (ICC) were used to estimate absolute agreement between raters and/or algorithms.ResultsAbsolute agreement between raters was 'excellent' for MRS (ICC = 0.97; 95%CI [0.96, 0.98]) and 'moderate' to 'good' for CPS (ICC = 0.77; 95%CI [0.69, 0.83]). For CPS, inter-rater reliability was poorer among less experienced raters (ICC = 0.70; 95%CI [0.57, 0.80]) when compared to experienced ones (ICC = 0.82; 95%CI [0.76, 0.88]). Absolute agreement between the two algorithms was 'excellent' for MRS (ICC = 0.96; 95%CI [0.91, 0.98]). For CPS, the best possible agreement was found for CPS defined as the print size sustaining 80% of MRS (ICC = 0.77; 95%CI [0.68, 0.84]). Absolute agreement between raters and automated methods was 'excellent' for MRS (ICC = 0.96; 95% CI [0.88, 0.98] for SDev; ICC = 0.97; 95% CI [0.95, 0.98] for NLME). For CPS, absolute agreement between raters and SDev ranged from 'poor' to 'good' (ICC = 0.66; 95% CI [0.3, 0.80]), while agreement between raters and NLME was 'good' (ICC = 0.83; 95% CI [0.76, 0.88]).ConclusionFor MRS, inter-rater reliability is excellent, even considering the possibility of noisy and/or incomplete data collected in low-vision individuals. For CPS, inter-rater reliability is lower. This may be problematic, for instance in the context of multisite investigations or follow-up examinations. The NLME method showed better agreement with the raters than the SDev method for both reading parameters. Setting up consensual guidelines to deal with ambiguous curves may help improve reliability. While the exact definition of CPS should be chosen on a case-by-case basis depending on the clinician or researcher's motivations, evidence suggests that estimating CPS as the smallest print size sustaining about 80% of MRS would increase inter-rater reliability.
| S-EPMC6555504 | biostudies-literature
Project description:Healthy young adults display a leftward asymmetry of spatial attention ("pseudoneglect") that has been measured with a wide range of different tasks. Yet at present there is a lack of systematic evidence that the tasks commonly used in research today are i) stable measures over time and ii) provide similar measures of spatial bias. Fifty right-handed young adults were tested on five tasks (manual line bisection, landmark, greyscales, gratingscales and lateralised visual detection) on two different days. All five tasks were found to be stable measures of bias over the two testing sessions, indicating that each is a reliable measure in itself. Surprisingly, no strongly significant inter-task correlations were found. However, principal component analysis revealed left-right asymmetries to be subdivided in 4 main components, namely asymmetries in size judgements (manual line bisection and landmark), luminance judgements (greyscales), stimulus detection (lateralised visual detection) and judgements of global/local features (manual line bisection and grating scales). The results align with recent research on hemispatial neglect which conceptualises the condition as multi-component rather than a single pathological deficit of spatial attention. We conclude that spatial biases in judgment of visual stimulus features in healthy adults (e.g., pseudoneglect) is also a multi-component phenomenon that may be captured by variations in task demands which engage task-dependent patterns of activation within the attention network.
| S-EPMC4574708 | biostudies-literature
Project description:BackgroundMusculoskeletal disorders particularly in the back and lower extremities are common among marines. Here, movement-control tests are considered clinically useful for screening and follow-up evaluation. However, few studies have addressed the reliability of clinical tests, and no such published data exists for marines. The present aim was therefore to determine the inter- and intra-observer reliability of clinically convenient tests emphasizing movement control of the back and hip among marines. A secondary aim was to investigate the sensitivity and specificity of these clinical tests for discriminating musculoskeletal pain disorders in this group of military personnel.MethodsThis inter- and intra-observer reliability study used a test-retest approach with six standardized clinical tests focusing on movement control for back and hip. Thirty-three marines (age 28.7 yrs, SD 5.9) on active duty volunteered and were recruited. They followed an in-vivo observation test procedure that covered both low- and high-load (threshold) tasks relevant for marines on operational duty. Two independent observers simultaneously rated performance as "correct" or "incorrect" following a standardized assessment protocol. Re-testing followed 7-10 days thereafter. Reliability was analysed using kappa (κ) coefficients, while discriminative power of the best-fitting tests for back- and lower-extremity pain was assessed using a multiple-variable regression model.ResultsInter-observer reliability for the six tests was moderate to almost perfect with κ-coefficients ranging between 0.56-0.95. Three tests reached almost perfect inter-observer reliability with mean κ-coefficients > 0.81. However, intra-observer reliability was fair-to-moderate with mean κ-coefficients between 0.22-0.58. Three tests achieved moderate intra-observer reliability with κ-coefficients > 0.41. Combinations of one low- and one high-threshold test best discriminated prior back pain, but results were inconsistent for lower-extremity pain.ConclusionsOur results suggest that clinical tests of movement control of back and hip are reliable for use in screening protocols using several observers with marines. However, test-retest reproducibility was less accurate, which should be considered in follow-up evaluations. The results also indicate that combinations of low- and high-threshold tests have discriminative validity for prior back pain, but were inconclusive for lower-extremity pain.
| S-EPMC3552977 | biostudies-literature
Project description:ObjectivesTo investigate the intra- and inter-rater reliability of the total radiomics quality score (RQS) and the reproducibility of individual RQS items' score in a large multireader study.MethodsNine raters with different backgrounds were randomly assigned to three groups based on their proficiency with RQS utilization: Groups 1 and 2 represented the inter-rater reliability groups with or without prior training in RQS, respectively; group 3 represented the intra-rater reliability group. Thirty-three original research papers on radiomics were evaluated by raters of groups 1 and 2. Of the 33 papers, 17 were evaluated twice with an interval of 1 month by raters of group 3. Intraclass coefficient (ICC) for continuous variables, and Fleiss' and Cohen's kappa (k) statistics for categorical variables were used.ResultsThe inter-rater reliability was poor to moderate for total RQS (ICC 0.30-055, p < 0.001) and very low to good for item's reproducibility (k - 0.12 to 0.75) within groups 1 and 2 for both inexperienced and experienced raters. The intra-rater reliability for total RQS was moderate for the less experienced rater (ICC 0.522, p = 0.009), whereas experienced raters showed excellent intra-rater reliability (ICC 0.91-0.99, p < 0.001) between the first and second read. Intra-rater reliability on RQS items' score reproducibility was higher and most of the items had moderate to good intra-rater reliability (k - 0.40 to 1).ConclusionsReproducibility of the total RQS and the score of individual RQS items is low. There is a need for a robust and reproducible assessment method to assess the quality of radiomics research.Clinical relevance statementThere is a need for reproducible scoring systems to improve quality of radiomics research and consecutively close the translational gap between research and clinical implementation.Key points• Radiomics quality score has been widely used for the evaluation of radiomics studies. • Although the intra-rater reliability was moderate to excellent, intra- and inter-rater reliability of total score and point-by-point scores were low with radiomics quality score. • A robust, easy-to-use scoring system is needed for the evaluation of radiomics research.
| S-EPMC10957586 | biostudies-literature
Project description:Study objectivesDevelopment of inter-database generalizable sleep staging algorithms represents a challenge due to increased data variability across different datasets. Sharing data between different centers is also a problem due to potential restrictions due to patient privacy protection. In this work, we describe a new deep learning approach for automatic sleep staging, and address its generalization capabilities on a wide range of public sleep staging databases. We also examine the suitability of a novel approach that uses an ensemble of individual local models and evaluate its impact on the resulting inter-database generalization performance.MethodsA general deep learning network architecture for automatic sleep staging is presented. Different preprocessing and architectural variant options are tested. The resulting prediction capabilities are evaluated and compared on a heterogeneous collection of six public sleep staging datasets. Validation is carried out in the context of independent local and external dataset generalization scenarios.ResultsBest results were achieved using the CNN_LSTM_5 neural network variant. Average prediction capabilities on independent local testing sets achieved 0.80 kappa score. When individual local models predict data from external datasets, average kappa score decreases to 0.54. Using the proposed ensemble-based approach, average kappa performance on the external dataset prediction scenario increases to 0.62. To our knowledge this is the largest study by the number of datasets so far on validating the generalization capabilities of an automatic sleep staging algorithm using external databases.ConclusionsValidation results show good general performance of our method, as compared with the expected levels of human agreement, as well as to state-of-the-art automatic sleep staging methods. The proposed ensemble-based approach enables flexible and scalable design, allowing dynamic integration of local models into the final ensemble, preserving data locality, and increasing generalization capabilities of the resulting system at the same time.
| S-EPMC8366993 | biostudies-literature