scholarly journals Subcortical brain segmentation in 5-year-old children: validation of FSL-FIRST and FreeSurfer against manual segmentation

2021 ◽  
Author(s):  
Kristian Lidauer ◽  
Elmo P Pulli ◽  
Anni Copeland ◽  
Eero Silver ◽  
Venla Kumpulainen ◽  
...  
Keyword(s):  
Nucleus Accumbens ◽  
Globus Pallidus ◽  
Pearson Correlation ◽  
Intraclass Correlation ◽  
Automated Segmentation ◽  
Manual Segmentation ◽  
Crucial Issue ◽  
Volumetric Quantification ◽  
Subcortical Nuclei ◽  
Automated Methods

Developing accurate subcortical volumetric quantification tools is a crucial issue for neurodevelopmental studies, as they could reduce the need for challenging and time-consuming manual segmentation. In this study the accuracy of two automated segmentation tools, FSL-FIRST (with three different boundary correction settings) and FreeSurfer were compared against manual segmentation of subcortical nuclei, including the hippocampus, amygdala, thalamus, putamen, globus pallidus, caudate and nucleus accumbens, using volumetric and correlation analyses in 80 5-year-olds. Both FSL-FIRST and FreeSurfer overestimated the volume on all structures except the caudate, and the accuracy depended on the structure. Small structures such as the amygdala and nucleus accumbens, which are visually difficult to distinguish, produced considerable overestimations and weaker correlations with all automated methods. Larger and more readily distinguishable structures such as the caudate and putamen produced notably lower overestimations and stronger correlations. Overall, the segmentations performed by FSL-FIRST's Default pipeline were the most accurate, while FreeSurfer's results were weaker across the structures. In line with prior studies, the accuracy of automated segmentation tools was imperfect with respect to manually defined structures. However, apart from amygdala and nucleus accumbens, FSL-FIRST's agreement could be considered satisfactory (Pearson correlation > 0.74, Intraclass correlation coefficient (ICC) > 0.68 and Dice Score coefficient (DSC) > 0.87) with highest values for the striatal structures (putamen, globus pallidus and caudate) (Pearson correlation > 0.77, ICC > 0.87 and DSC > 0.88, respectively). Overall, automated segmentation tools do not always provide satisfactory results, and careful visual inspection of the automated segmentations is strongly advised.

scholarly journals The Patient’s Denture Assessment (Thai version) is a valid and reliable tool for evaluating the outcome of treatment with complete denture

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Sahaprom Namano ◽  
Orapin Komin
Keyword(s):  
Treatment Outcome ◽  
Pearson Correlation ◽  
Treatment Success ◽  
Intraclass Correlation ◽  
Reliability And Validity ◽  
Complete Denture ◽  
Care Clinic ◽  
Geriatric Dentistry ◽  
Significant Difference ◽  
Denture Treatment

Abstract Background Complete tooth losses are still being major problems which resulted in lesser quality of life especially for elderly patients. However, there are still lack of questionnaire to evaluate the treatment outcome from the patient’s aspect. The objective of this study is to evaluate the reliability and validity of the Patient’s Denture Assessment-Thai version (PDA-T), then use this questionnaire to assess patient satisfaction with complete denture treatment outcome also investigates the factors involving their satisfaction. Methods The subjects comprised 120 edentulous adult patients (49 men/71 women; average age 70 years-old) from the Prosthodontic and the Geriatric Dentistry and Special Patients Care Clinic at the Faculty of Dentistry, Chulalongkorn University during 2019 March‒2020 March. The patients were divided into two groups: the group experienced (Exper) (n = 54) with wearing complete dentures, and the non-experienced (NonExper) group (n = 66). The patients used the validated PDA-T to self-assess their treatment at different times. The Exper group completed the questionnaire at t0 (during treatment), t0.5 (2‒8-weeks after t0), and t1 (final follow-up). The NonExper group completed the questionnaire only at t1. Results In the Exper group, Cronbach’s α and average inter-item correlation was 0.95 (range 0.76‒0.95) and 0.47 (range 0.57‒0.83), respectively. The intraclass correlation coefficients (n = 18, 95% confidence interval) were 0.98 overall. The paired t-test (p < 0.05) between t0 and t1 indicated a significant difference between t0 and t1 in every PDA-T topic, and the effect size was 1.71. In the NonExper group, the Pearson correlation analysis indicated no significant correlation between the patients' demographics and masticatory function. Conclusion The reliability and validity of the PDA-T indicate it is a valuable tool for evaluating complete denture treatment. Treatment success affected the patients' satisfaction but was not associated with the type of doctors, genders, ages, or educational level.

Abstract WP406: Training and Validation of Deepmedic Machine Learning Tool for Automated Hematoma Segmentation and Volume Analysis on CT Using Multicenter Data

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Sean Nurmsoo ◽  
Alessandro Guida ◽  
Alex Wong ◽  
Richard I Aviv ◽  
Andrew Demchuk ◽  
...  
Keyword(s):  
Machine Learning ◽  
Validation Cohort ◽  
Learning Algorithm ◽  
Intraclass Correlation ◽  
Ct Scans ◽  
Dice Similarity Coefficient ◽  
Machine Learning Algorithm ◽  
Manual Segmentation ◽  
Training Cohort ◽  
Volume Calculation

Introduction: We sought to train and validate an automated machine learning algorithm for ICH segmentation and volume calculation using multicenter data. Methods: An open-source 3D deep machine learning algorithm “DeepMedic” was trained using manually segmented ICH from 208 CT scans (129 patients) from the multicenter PREDICT study. The algorithm was then validated with 125 manually segmented CT scans (48 patients) from the SPOTLIGHT study. Manual segmentation was performed with Quantomo semi-automated software. ABC/2 was measured for all studies by two neuroradiologists. Accuracy of DeepMedic segmentation was assessed using the Dice similarity coefficient. Analysis was stratified by presence of IVH. Intraclass correlation (ICC) with 95% confidence intervals (CI) assessed agreement between manual vs. DeepMedic segmentation volume; and manual segmentation and ABC/2 volume. Bland-Altman charts were analyzed for ABC/2 and DeepMedic vs. manual segmentation volumes. Results: DeepMedic demonstrated high segmentation accuracy in the training cohort (median Dice 0.96; IQR 0.95 - 0.97) and in the validation cohort (median Dice 0.91; IQR 0.86 - 0.94). Dice coefficients were not significantly different between patients with IVH in the training cohort; however was significantly worse in the validation cohort in patients with IVH (Wilcoxon p<0.001). Agreement was significantly better between DeepMedic and manual segmentation (PREDICT: ICC 0.99 [95%CI 0.99 -1.00]; SPOTLIGHT: ICC 0.98 [95%CI 0.97 - 0.99]) than between ABC/2 and manual segmentation (PREDICT: ICC 0.92 [95%CI 0.89 - 0.95]; SPOTLIGHT: ICC 0.95 [95%CI 0.93-0.97]). Improved accuracy of DeepMedic was demonstrated in Bland-Altman charts (Fig 1). Conclusion: ICH machine learning segmentation with DeepMedic is feasible and accurate; and demonstrates greater agreement with manual segmentation compared to ABC/2 volumes. Accuracy of the machine learning algorithm however is limited in patients with IVH.

scholarly journals Adaptation and validation of the Caregiver Burden Inventory for use with caregivers of elderly individuals

2015 ◽  
Vol 23 (1) ◽  
pp. 130-138 ◽  
Author(s):  
Daiany Borghetti Valer ◽  
Marinês Aires ◽  
Fernanda Lais Fengler ◽  
Lisiane Manganelli Girardi Paskulin
Keyword(s):  
Older Adults ◽  
Correlation Coefficient ◽  
Caregiver Burden ◽  
Pearson Correlation ◽  
Intraclass Correlation ◽  
Expert Committee ◽  
Five Dimensions ◽  
Social And Emotional ◽  
Back Translation ◽  
Student’S T

OBJECTIVE: to adapt and validate the Caregiver Burden Inventory for use with caregivers of older adults in Brazil.METHOD: methodological study involving initial translation, synthesis of translations, back translation, expert committee review, pre-testing, submission of the final version to the original authors, and assessment of the inventory's psychometric properties. The inventory assesses five dimensions of caregiver burden: time-dependence, developmental, physical, social and emotional dimensions.RESULTS: a total of 120 family caregivers took part in the study. All care-receivers were older adults dependent on assistance to perform activities of daily living, and lived in the central region of the city of Porto Alegre, RS, Brasil. Cronbach's alpha value for the inventory was 0.936, and the Pearson correlation coefficient for the relationship between the scores obtained on the Caregiver Burden Inventory and the Burden Interview was 0.814. The intraclass correlation coefficient was 0.941, and the value of Student's T-test comparing test and retest scores was 0.792.CONCLUSION: the instrument presented adequate reliability and the suitability of its items and factors was confirmed in this study.

Segmentation of design protocol using EEG

2018 ◽  
Vol 33 (1) ◽  
pp. 11-23 ◽  
Author(s):  
Philon Nguyen ◽  
Thanh An Nguyen ◽  
Yong Zeng
Keyword(s):  
Automated Segmentation ◽  
Manual Segmentation ◽  
Engineering Systems ◽  
Verbal Protocols ◽  
Segmentation Method ◽  
Average Deviation ◽  
Domain Experts ◽  
Design Activities ◽  
The Mind ◽  
Design Protocol

AbstractDesign protocol data analysis methods form a well-known set of techniques used by design researchers to further understand the conceptual design process. Verbal protocols are a popular technique used to analyze design activities. However, verbal protocols are known to have some limitations. A recurring problem in design protocol analysis is to segment and code protocol data into logical and semantic units. This is usually a manual step and little work has been done on fully automated segmentation techniques. Physiological signals such as electroencephalograms (EEG) can provide assistance in solving this problem. Such problems are typical inverse problems that occur in the line of research. A thought process needs to be reconstructed from its output, an EEG signal. We propose an EEG-based method for design protocol coding and segmentation. We provide experimental validation of our methods and compare manual segmentation by domain experts to algorithmic segmentation using EEG. The best performing automated segmentation method (when manual segmentation is the baseline) is found to have an average deviation from manual segmentations of 2 s. Furthermore, EEG-based segmentation can identify cognitive structures that simple observation of design protocols cannot. EEG-based segmentation does not replace complex domain expert segmentation but rather complements it. Techniques such as verbal protocols are known to fail in some circumstances. EEG-based segmentation has the added feature that it is fully automated and can be readily integrated in engineering systems and subsystems. It is effectively a window into the mind.

scholarly journals A comparative analysis of automated MRI brain segmentation in a large longitudinal dataset: Freesurfer vs. FSL

2020 ◽  
Author(s):  
Javier Quilis-Sancho ◽  
Miguel A. Fernandez-Blazquez ◽  
J Gomez-Ramirez
Keyword(s):  
Comparative Analysis ◽  
Brain Structures ◽  
Automated Segmentation ◽  
Manual Segmentation ◽  
Brain Segmentation ◽  
Clinical Tool ◽  
Brain Volumetry ◽  
Subcortical Brain Structures ◽  
Volume Estimates ◽  
Longitudinal Dataset

AbstractThe study of brain volumetry and morphology of the different brain structures can determine the diagnosis of an existing disease, quantify its prognosis or even help to identify an early detection of dementia. Manual segmentation is an extremely time consuming task and automated methods are thus, gaining importance as clinical tool for diagnosis. In the last few years, AI-based segmentation has delivered, in some cases, superior results than manual segmentation, in both time and accuracy. In this study we aim at performing a comparative analysis of automated brain segmentation. In order to test the performance of automated segmentation methods, the two most commonly used software libraries for brain segmentation Freesurfer and FSL, were put to work in each of the 4028 MRIs available in the study. We find a lack of linear correlation between the segmentation results obtained from Freesurfer and FSL. On the other hand. Freesurfer volume estimates of subcortical brain structures tends to be larger than FSL estimates of same areas. The study builds on an uniquely large, longitudinal dataset of over 4,000 MRIs, all performed with identical equipment to help researchers understand what to expect from fully automated segmentation procedures.

scholarly journals A Quantitative Comparison of Multispectral Refraction Topography and Autorefractometer in Young Adults

2021 ◽  
Vol 8 ◽  
Author(s):  
Yunru Liao ◽  
Zhenlan Yang ◽  
Zijing Li ◽  
Rui Zeng ◽  
Jing Wang ◽  
...  
Keyword(s):  
Correlation Coefficient ◽  
Pearson Correlation ◽  
Intraclass Correlation ◽  
Correlation Coefficients ◽  
Cross Sectional Study ◽  
Cross Sectional ◽  
High Agreement ◽  
Intraclass Correlation Coefficients ◽  
Repeatability And Reproducibility ◽  
The Mean

Purpose: Purpose of this study is to evaluate the measuring consistency of central refraction between multispectral refraction topography (MRT) and autorefractometry.Methods: This was a descriptive cross-sectional study including subjects in Sun Yat-sen Memorial Hospital from September 1, 2020, to December 31, 2020, ages 20 to 35 years with a best corrected visual acuity of 20/20 or better. All patients underwent cycloplegia, and the refractive status was estimated with autorefractometer, experienced optometrist and MRT. We analyzed the central refraction of the autorefractometer and MRT. The repeatability and reproducibility of values measured using both devices were evaluated using intraclass correlation coefficients (ICCs).Results: A total of 145 subjects ages 20 to 35 (290 eyes) were enrolled. The mean central refraction of the autorefractometer was −4.69 ± 2.64 diopters (D) (range −9.50 to +4.75 D), while the mean central refraction of MRT was −4.49 ± 2.61 diopters (D) (range −8.79 to +5.02 D). Pearson correlation analysis revealed a high correlation between the two devices. The intraclass correlation coefficient (ICC) also showed high agreement. The intrarater and interrater ICC values of central refraction were more than 0.90 in both devices and conditions. At the same time, the mean central refraction of experienced optometrist was −4.74 ± 2.66 diopters (D) (range −9.50 to +4.75D). The intra-class correlation coefficient of central refraction measured by MRT and subjective refraction was 0.939.Conclusions: Results revealed that autorefractometry, experienced optometrist and MRT show high agreement in measuring central refraction. MRT could provide a potential objective method to assess peripheral refraction.

The Reliability of Measuring Early Brain Edema with Computed Tomography

2021 ◽  
Vol 17 ◽  
Author(s):  
Yi-Fang Fan ◽  
Mi Shen ◽  
Xin-Xin Wang ◽  
Xiao-Yuan Liu ◽  
Yu-Ming Peng ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Brain Edema ◽  
Early Stage ◽  
Pearson Correlation ◽  
Intraclass Correlation ◽  
Region Of Interest ◽  
High Grade Glioma ◽  
High Grade ◽  
Magnetic Resonance Imaging Mri ◽  
Ct And Mri

Background: Postoperative brain edema is a common complication in patients with high-grade glioma after craniotomy. Both computed tomography (CT) and Magnetic Resonance Imaging (MRI) are applied to diagnose brain edema. Usually, MRI is considered to be better than CT for identifying brain edema. However, MRI is not generally applied in diagnosing acute cerebral edema in the early postoperative stage. Whether CT is reliable in detecting postoperative brain edema in the early stage is unknown. Objective: To investigate the agreement and correlation between CT and MRI for measuring early postoperative brain edema. Methods: Patients with high-grade glioma who underwent craniotomy in Beijing Tiantan hospital from January 2017 to October 2018 were retrospectively analyzed. The region of interest and operative cavity were manually outlined, and the volume of postoperative brain edema was measured on CT and MRI. Pearson correlation testing and the intraclass correlation coefficient (ICC) were used to evaluate the association and agreement between CT and MRI for detecting the volume of postoperative brain edema. Results: Twenty patients were included in this study. The interrater agreement was perfect for detecting brain edema (CT: κ=1, ICC=0.977, P<0.001; MRI: κ=0.866, ICC=0.963, P<0.001). A significant positive correlation and excellent consistency between CT and MRI were found for measuring the volume of brain edema (rater 1: r=0.97, ICC=0.934, P<0.001; rater 2: r=0.97, ICC=0.957, P<0.001). Conclusion: Substantial comparability between CT and MRI is demonstrated for detecting postoperative brain edema. It is reliable to use CT for measuring brain edema volume in the early stage after surgery.

754 STEEP RAMP TEST: AN ALTERNATIVE TO ASSESS PHYSICAL FITNESS BEFORE OESOPHAGECTOMY IN CANCER PATIENTS

2021 ◽  
Vol 34 (Supplement_1) ◽  
Author(s):  
David Crull ◽  
I Mekenkamp ◽  
G M F Ruinemans ◽  
M J Det ◽  
E A Kouwenhoven
Keyword(s):  
Physical Fitness ◽  
Cancer Patients ◽  
Oesophageal Cancer ◽  
Perioperative Complications ◽  
Pearson Correlation ◽  
Intraclass Correlation ◽  
Cardiopulmonary Exercise Test ◽  
Promising Alternative ◽  
Ramp Test ◽  
The Difference

Abstract   Oesophagectomy patients are at high risk of perioperative complications. An important predictor for a complication is inadequate preoperative physical fitness. Cardiopulmonary Exercise Test (CPET) is the golden standard for measuring VO₂ max. The Steep Ramp Test (SRT) is an alternative with a lower patient burden and cost, but provides an estimation of VO₂ max. The aim was to determine whether SRT is an adequate alternative for CPET to determine preoperative fitness in oesophageal cancer patients. Methods The population consisted of 113 oesophageal cancer patients of a single centre that have performed CPET and SRT within a timeframe of two weeks. The agreement between SRT and CPET was analysed using a t-test, the Pearson correlation, the Intraclass Correlation Coefficient (ICC) and the Bland-Altmann analysis. A CPET-measured VO₂ max of 17 mL/kg/min was the set threshold for adequate operative fitness, based on current literature. Results The mean difference between the CPET and SRT was 2.77 mL/kg/min (95% CI, 2.14–3.41). The Pearson correlation was 0.792 (P &lt; 0.001). The ICC was 0.879 (95% CI, 0.825–0.917). In 93 (82.3%) patients, SRT was higher than the CPET. The limits of agreement (LOA) for the Bland-Altmann were − 3.89 − +9.44. Thirty-one (27.4%) patients performed better on SRT than the set threshold plus the upper LOA. Twenty-three patients scored inadequate on CPET, of whom 14 patients (60.9%) SRT was sufficient. Conclusion VO₂ max as estimated by SRT differs from the VO₂ max as measured by CPET. In patients that are well above the threshold of adequate preoperative fitness, the difference is of clinical irrelevance. In conclusion: SRT is a promising alternative to CPET for determining preoperative physical fitness, it might render CPET obsolete for fit individuals which leads to a lower burden for the patient and substantial cost reduction.

IVUS Image Processing Methodologies

2013 ◽  
pp. 639-657
Author(s):  
Antonis A. Sakellarios ◽  
Christos V. Bourantas ◽  
Lambros S. Athanasiou ◽  
Dimitrios I. Fotiadis ◽  
Lampros K. Michalis
Keyword(s):  
Image Processing ◽  
Vessel Wall ◽  
Regions Of Interest ◽  
Manual Segmentation ◽  
Arterial Lumen ◽  
Ivus Image ◽  
Processing Techniques ◽  
Automated Methods

Intravascular Ultrasound (IVUS) is an invasive imaging technique that allows detailed visualization of the arterial lumen and outer vessel wall and permits characterization of the type of the plaque and quantification of its burden. Traditionally IVUS processing was performed manually. However, it became apparent that manual segmentation is time consuming, and the obtained results depend on the experience of the operators. To overcome these limitations and enhance the role of IVUS in clinical practice and research, several (semi-) automated methods have been developed that expedite detection of the regions of interest and/or characterization of the type of the plaque. In this chapter we review the available IVUS processing techniques and present the developed commercial solutions for IVUS segmentation and plaque characterization.

scholarly journals LUMINOUS database: lumbar multifidus muscle segmentation from ultrasound images

2020 ◽  
Vol 21 (1) ◽  
Author(s):  
Clyde J. Belasso ◽  
Bahareh Behboodi ◽  
Habib Benali ◽  
Mathieu Boily ◽  
Hassan Rivaz ◽  
...  
Keyword(s):  
Ground Truth ◽  
Ease Of Use ◽  
Automated Segmentation ◽  
Manual Segmentation ◽  
Clinical Implementation ◽  
Multifidus Muscle ◽  
Lumbar Multifidus ◽  
The Us ◽  
Segmentation Algorithms ◽  
And Function

Abstract Background Among the paraspinal muscles, the structure and function of the lumbar multifidus (LM) has become of great interest to researchers and clinicians involved in lower back pain and muscle rehabilitation. Ultrasound (US) imaging of the LM muscle is a useful clinical tool which can be used in the assessment of muscle morphology and function. US is widely used due to its portability, cost-effectiveness, and ease-of-use. In order to assess muscle function, quantitative information of the LM must be extracted from the US image by means of manual segmentation. However, manual segmentation requires a higher level of training and experience and is characterized by a level of difficulty and subjectivity associated with image interpretation. Thus, the development of automated segmentation methods is warranted and would strongly benefit clinicians and researchers. The aim of this study is to provide a database which will contribute to the development of automated segmentation algorithms of the LM. Construction and content This database provides the US ground truth of the left and right LM muscles at the L5 level (in prone and standing positions) of 109 young athletic adults involved in Concordia University’s varsity teams. The LUMINOUS database contains the US images with their corresponding manually segmented binary masks, serving as the ground truth. The purpose of the database is to enable development and validation of deep learning algorithms used for automatic segmentation tasks related to the assessment of the LM cross-sectional area (CSA) and echo intensity (EI). The LUMINOUS database is publicly available at http://data.sonography.ai. Conclusion The development of automated segmentation algorithms based on this database will promote the standardization of LM measurements and facilitate comparison among studies. Moreover, it can accelerate the clinical implementation of quantitative muscle assessment in clinical and research settings.

Sign in / Sign up

Export Citation Format

Share Document