scholarly journals COVLIAS 1.0 vs. MedSeg: Artificial Intelligence-Based Comparative Study for Automated COVID-19 Computed Tomography Lung Segmentation in Italian and Croatian Cohorts

Diagnostics ◽  
2021 ◽  
Vol 11 (12) ◽  
pp. 2367
Author(s):  
Jasjit S. Suri ◽  
Sushant Agarwal ◽  
Alessandro Carriero ◽  
Alessio Paschè ◽  
Pietro S. C. Danna ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Experimental Data ◽  
Computed Tomography ◽  
Characteristic Curve ◽  
Computing Time ◽  
Mean Value ◽  
Image Analysis System ◽  
Validation Data ◽  
Lung Segmentation ◽  
The Mean

(1) Background: COVID-19 computed tomography (CT) lung segmentation is critical for COVID lung severity diagnosis. Earlier proposed approaches during 2020–2021 were semiautomated or automated but not accurate, user-friendly, and industry-standard benchmarked. The proposed study compared the COVID Lung Image Analysis System, COVLIAS 1.0 (GBTI, Inc., and AtheroPointTM, Roseville, CA, USA, referred to as COVLIAS), against MedSeg, a web-based Artificial Intelligence (AI) segmentation tool, where COVLIAS uses hybrid deep learning (HDL) models for CT lung segmentation. (2) Materials and Methods: The proposed study used 5000 ITALIAN COVID-19 positive CT lung images collected from 72 patients (experimental data) that confirmed the reverse transcription-polymerase chain reaction (RT-PCR) test. Two hybrid AI models from the COVLIAS system, namely, VGG-SegNet (HDL 1) and ResNet-SegNet (HDL 2), were used to segment the CT lungs. As part of the results, we compared both COVLIAS and MedSeg against two manual delineations (MD 1 and MD 2) using (i) Bland–Altman plots, (ii) Correlation coefficient (CC) plots, (iii) Receiver operating characteristic curve, and (iv) Figure of Merit and (v) visual overlays. A cohort of 500 CROATIA COVID-19 positive CT lung images (validation data) was used. A previously trained COVLIAS model was directly applied to the validation data (as part of Unseen-AI) to segment the CT lungs and compare them against MedSeg. (3) Result: For the experimental data, the four CCs between COVLIAS (HDL 1) vs. MD 1, COVLIAS (HDL 1) vs. MD 2, COVLIAS (HDL 2) vs. MD 1, and COVLIAS (HDL 2) vs. MD 2 were 0.96, 0.96, 0.96, and 0.96, respectively. The mean value of the COVLIAS system for the above four readings was 0.96. CC between MedSeg vs. MD 1 and MedSeg vs. MD 2 was 0.98 and 0.98, respectively. Both had a mean value of 0.98. On the validation data, the CC between COVLIAS (HDL 1) vs. MedSeg and COVLIAS (HDL 2) vs. MedSeg was 0.98 and 0.99, respectively. For the experimental data, the difference between the mean values for COVLIAS and MedSeg showed a difference of <2.5%, meeting the standard of equivalence. The average running times for COVLIAS and MedSeg on a single lung CT slice were ~4 s and ~10 s, respectively. (4) Conclusions: The performances of COVLIAS and MedSeg were similar. However, COVLIAS showed improved computing time over MedSeg.

scholarly journals COVLIAS 1.0: Lung Segmentation in COVID-19 Computed Tomography Scans Using Hybrid Deep Learning Artificial Intelligence Models

Diagnostics ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 1405
Author(s):  
Jasjit S. Suri ◽  
Sushant Agarwal ◽  
Rajesh Pathak ◽  
Vedmanvitha Ketireddy ◽  
Marta Columbu ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
Ground Truth ◽  
P Value ◽  
Image Analysis System ◽  
Validation Data ◽  
Lung Segmentation ◽  
Intelligence Models ◽  
Artificial Intelligence Models ◽  
Analysis System

Background: COVID-19 lung segmentation using Computed Tomography (CT) scans is important for the diagnosis of lung severity. The process of automated lung segmentation is challenging due to (a) CT radiation dosage and (b) ground-glass opacities caused by COVID-19. The lung segmentation methodologies proposed in 2020 were semi- or automated but not reliable, accurate, and user-friendly. The proposed study presents a COVID Lung Image Analysis System (COVLIAS 1.0, AtheroPoint™, Roseville, CA, USA) consisting of hybrid deep learning (HDL) models for lung segmentation. Methodology: The COVLIAS 1.0 consists of three methods based on solo deep learning (SDL) or hybrid deep learning (HDL). SegNet is proposed in the SDL category while VGG-SegNet and ResNet-SegNet are designed under the HDL paradigm. The three proposed AI approaches were benchmarked against the National Institute of Health (NIH)-based conventional segmentation model using fuzzy-connectedness. A cross-validation protocol with a 40:60 ratio between training and testing was designed, with 10% validation data. The ground truth (GT) was manually traced by a radiologist trained personnel. For performance evaluation, nine different criteria were selected to perform the evaluation of SDL or HDL lung segmentation regions and lungs long axis against GT. Results: Using the database of 5000 chest CT images (from 72 patients), COVLIAS 1.0 yielded AUC of ~0.96, ~0.97, ~0.98, and ~0.96 (p-value < 0.001), respectively within 5% range of GT area, for SegNet, VGG-SegNet, ResNet-SegNet, and NIH. The mean Figure of Merit using four models (left and right lung) was above 94%. On benchmarking against the National Institute of Health (NIH) segmentation method, the proposed model demonstrated a 58% and 44% improvement in ResNet-SegNet, 52% and 36% improvement in VGG-SegNet for lung area, and lung long axis, respectively. The PE statistics performance was in the following order: ResNet-SegNet > VGG-SegNet > NIH > SegNet. The HDL runs in <1 s on test data per image. Conclusions: The COVLIAS 1.0 system can be applied in real-time for radiology-based clinical settings.

scholarly journals Normalization in dynamic speckle analysis for non-destructive monitoring of speed of processes

2021 ◽  
Vol 2091 (1) ◽  
pp. 012002
Author(s):  
Elena Stoykova ◽  
Dimana Nazarova ◽  
Lian Nedelchev ◽  
Mikhail Levchenko ◽  
Nataliya Berberova-Buhova ◽  
...  
Keyword(s):  
Experimental Data ◽  
Mean Value ◽  
High Contrast ◽  
Object Surface ◽  
Laser Illumination ◽  
Dynamic Speckle ◽  
The Mean ◽  
Spatially Varying ◽  
Non Destructive

Abstract The paper is dedicated to analysis of normalized intensity-based pointwise algorithms for processing dynamic speckle images with spatially varying speckle statistics in non-destructive visualization of regions of faster or slower changes across an object. Both existing and newly proposed algorithms are analyzed. Extraction of speed of changes is done by acquiring correlated in time speckle images formed on the object surface under laser illumination. The studied algorithms have been applied to simulated low and high contrast speckle data. Their performance has been compared to processing of binary patterns as another approach for dealing with varying speckle statistics in the acquired images. The efficiency of the algorithms have been checked on the experimental data, including data in a compressed format. We have proven that the algorithms with normalization at successive instants by a sum of two intensities or a single intensity outperform as a whole the algorithms which apply the time-averaged estimates of the mean value and the variance of speckle intensity.

Neutrophil to lymphocyte ratio as a predictive biomarker for computed tomography scan use in mild traumatic brain injury

2020 ◽  
Vol 14 (12) ◽  
pp. 1085-1090
Author(s):  
George A Alexiou ◽  
Georgios D Lianos ◽  
Aggeliki Tzima ◽  
Athanasios Sotiropoulos ◽  
Anastasios Nasios ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Traumatic Brain Injury ◽  
Brain Injury ◽  
Characteristic Curve ◽  
Predictive Biomarker ◽  
Health Concern ◽  
Mild Tbi ◽  
Ct Findings ◽  
The Mean

Aim: Traumatic brain injury (TBI) is a serious health concern. We set out to investigate the role of neutrophil-to-lymphocytes ratio (NLR) at admission for predicting the need for computed tomography (CT) in mild-TBI. Materials & methods: A retrospective study of adult patients who presented with mild-TBI Results: One hundred and thirty patients met the inclusion criteria. Seventy-four patients had positive CT-findings. The mean NLR-levels at presentations were 5.6 ± 4.8. Patients with positive CT-findings had significant higher NLR-levels. Receiver operating characteristic curve analysis was conducted and the threshold of NLR-levels for detecting the cases with positive CT-findings was 2.5, with 78.1% sensitivity and 63% specificity Conclusion: To the best of our knowledge no previous study has assessed the value of NLR-levels for predicting the need for CT in mild-TBI.

scholarly journals Articular Eminence Inclination, Height, and Condyle Morphology on Cone Beam Computed Tomography

2014 ◽  
Vol 2014 ◽  
pp. 1-6 ◽  
Author(s):  
Dilhan İlgüy ◽  
Mehmet İlgüy ◽  
Erdoğan Fişekçioğlu ◽  
Semanur Dölekoğlu ◽  
Nilüfer Ersan
Keyword(s):  
Computed Tomography ◽  
Cone Beam Computed Tomography ◽  
Mean Value ◽  
Cone Beam ◽  
Articular Eminence ◽  
Age And Gender ◽  
Mean Values ◽  
The Mean ◽  
And Gender ◽  
The Relationship

Aim. The aim of the present study was to examine the relationship between articular eminence inclination, height, and thickness of the roof of the glenoid fossa (RGF) according to age and gender and to assess condyle morphology including incidental findings of osseous characteristics associated with osteoarthritis (OA) of the temporomandibular joint (TMJ) using cone beam computed tomography (CBCT).Materials and Methods. CBCT images of 105 patients were evaluated retrospectively. For articular eminence inclination and height, axial views on which the condylar processes were seen with their widest mediolateral extent being used as a reference view for secondary reconstruction. Condyle morphology was categorized both in the sagittal and coronal plane.Results. The mean values of eminence inclination and height of males were higher than those of females (P<0.05). There were significant differences in the RGF thickness in relation to sagittal condyle morphology. Among the group of OA, the mean value of the RGF thickness for “OA-osteophyte” group was the highest (1.59 mm), whereas the lowest RGF values were seen in the “OA-flattening.”Conclusion. The sagittal osteoarthritic changes may have an effect on RGF thickness by mechanical stimulation and changed stress distribution. Gender has a significant effect on eminence height (Eh) and inclination.

Studies on the Anatomical Structure of Fast-Growing Toona ciliate

2012 ◽  
Vol 472-475 ◽  
pp. 1963-1966
Author(s):  
Jin Sun ◽  
Xiao Bo Wang ◽  
Xiao Jing Wang ◽  
Yan Lin ◽  
Zhen Zhong Gao
Keyword(s):  
Growth Ring ◽  
Mean Value ◽  
Image Analysis System ◽  
Fast Growing ◽  
Coarse Texture ◽  
Vessel Elements ◽  
Toona Ciliata ◽  
The Mean ◽  
Analysis System ◽  
Tissue Proportion

In this paper, anatomical characters and tissue proportion of fast-growing Toona ciliata samples from Guangzhou was investigated assisted with the light microscope and image analysis system based on the principle of stereology. Qualitative and quantitative data were presented for purposes of identification. Results showed that fast growing T. ciliata was coarse texture with wide growth ring of 3~5cm.The ray proportion was 12.12%-25.56%, the mean value was 17.98%, increasing with time-dependent relevance. The vessel-elements proportion was 6.37%-10.00%, the mean value was 7.40%, increasing then declining and finally increasing. The fiber and other tissue proportion was 67.06%-79.92%, the mean value was 74.62%, declining by year.

Plastic Energy Required for Fatigue Failure of Specimens Containing Cracks

1975 ◽  
Vol 97 (2) ◽  
pp. 172-178 ◽  
Author(s):  
Thang Bui-Quoc ◽  
Andre Biron
Keyword(s):  
Experimental Data ◽  
Crack Propagation ◽  
Hysteresis Loops ◽  
Mean Value ◽  
Repeated Loading ◽  
Plastic Energy ◽  
Propagation Law ◽  
The Mean ◽  
Damage Rule ◽  
Propagation Stage

A study is conducted to determine the plastic energy absorbed by a material subjected to cyclic loading during the crack propagation stage. The analysis is based on Liu’s elastoplastic solution for calculating the plastic energy around the crack tip combined with a modified crack propagation law. The plastic energy per cycle Δw is found to be dependent upon the fatigue crack length and therefore varies with the number of applied cycles. The mean value of this energy corresponds to the plastic energy obtained by measuring the area within the stress-strain hysteresis loops of the material under repeated loading. If fatigue damage is based on plastic energy, the trend of variation of Δw with the applied cycles in the present analysis shows that the application of the linear damage rule (such as Miner’s law) should be more successful in the low-cycle region than in the high-cycle region. This is consistent with experimental data.

scholarly journals Soft tissue changes after anterior maxillary osteotomy by using 3D CBCT scanning.

2018 ◽  
Vol 4 (4) ◽  
pp. 1245-1250 ◽  
Author(s):  
Ahmed AL-Omar ◽  
Usama AL-Dakroroy
Keyword(s):  
Computed Tomography ◽  
Soft Tissue ◽  
Cone Beam Computed Tomography ◽  
Soft Tissues ◽  
Mean Value ◽  
Cone Beam ◽  
Maxillary Osteotomy ◽  
Tissue Changes ◽  
The Mean ◽  
Soft Tissue Changes

The objective of this study was to evaluate the soft tissue changes after anterior maxillary osteotomy using cone beam computed tomography. The study included twelve patients diagnosed as suffering from an excess of the anterior part of maxilla and required surgical correction by anterior maxillary osteotomy. Assessment of Cone beam computed tomography in evaluation the soft tissue changes. All cases subjected to anterior maxillary osteotomy using modified Cuper technique. Follow up of cases was does clinically and radio- graphically by cone beam computed tomography. The results of the study proved that the statistically significant increase in mean Nasolabial angle after 6 months was 5.2 degrees, the mean value of nasal width after 6 months was 0,9mm. The mean value of Philtrum length was 3.3mm. In conclusion, Cone Beam Computed Tomography can be used to obtain dimensionally accurate linear and angular measurement from bony maxillofacial structure and landmarks and allows accurate, three-dimensional imaging of hard and soft tissues.

scholarly journals Root Canal Transportation after Root Canal Preparation with ProTaper Next, WaveOne Gold, and Twisted Files

2020 ◽  
Vol 9 (11) ◽  
pp. 3661
Author(s):  
Wojciech Eliasz ◽  
Kinga Kubiak ◽  
Wojciech Poncyljusz ◽  
Anna Surdacka
Keyword(s):  
Computed Tomography ◽  
Root Canal ◽  
Mean Value ◽  
Root Canal Preparation ◽  
Movement Kinematics ◽  
Root Canals ◽  
Adaptive Motion ◽  
Computed Tomography Scans ◽  
Canal Transportation ◽  
The Mean

Background: Root canal preparation during endodontic treatment may be associated with various complications, including a change in the original pathway of the root canal lumen. The aim of our study was to determine whether files of similar sizes that use various movement kinematics (rotary, reciprocal, adaptive motion) cause root canal transportation, and whether the differences between such systems are statistically significant. Methods: The degree of root canal transportation (DT) was calculated with the use of computed tomography scans for 3 groups of teeth (for each group: n = 20) in which the root canals were prepared using either rotary (ProTaper Next—PTN), reciprocal (WaveOne Gold—WOG), or adaptive movement (Twisted Files—TF) instruments. Results: For rotary ProTaper Next instruments, the mean value of the DT index was 0.0795 (SD = 0.0179) for 3 mm from the apex, 0.09 (SD = 0.0262) for 6 mm from the apex, and 0.106 (SD = 0.0221) for 9 mm from the apex. For reciprocal WaveOne Gold Primary instruments, the mean value of the DT index was 0.0355 (SD = 0.015) for 3 mm from the apex, 0.061 (SD = 0.02) for 6 mm from the apex, and 0.08 (SD = 0.25) for 9 mm from the apex. For Twisted Files, the mean value of the DT index was 0.05 (SD = 0.03) for 3 mm from the apex, 0.092 (SD = 0.17) for 6 mm from the apex, and 0.08 (SD = 0.02) for 9 mm from the apex. Conclusions: The use of PTN, WOG, and TF files resulted in root canal transportation to a different degree. The use of rotary PTN files produced the most transported preparation, whereas the use of WOG files produced the conservative root canal preparation that allowed the retention of the original shape of the root canal.

scholarly journals A Study on Measurements of Different Dimensions of Maxillary Sinus by Computed Tomography in Adult Nepalese Population

2018 ◽  
Vol 8 (2) ◽  
pp. 8-12
Author(s):  
Umesh Prasad Khanal ◽  
Anupama Adhikari ◽  
Mukunda Prasad Humagain
Keyword(s):  
Computed Tomography ◽  
Maxillary Sinus ◽  
Anterior Wall ◽  
Mean Value ◽  
Forensic Sciences ◽  
P Value ◽  
Male Population ◽  
Different Dimensions ◽  
Left And Right ◽  
The Mean

Introduction: Measurement of different dimensions of maxillary sinus and anterior wallthickness of maxillary sinus by Computed Tomography in normal Nepalese populationMethods: Dimensions of 90 patients were measured in CT PNS using Syngovia Software. AP diameter, width and anterior wall thickness were measured in axial images and height was measured in coronal images.Results: The mean volume of maxillary sinuses in study of male population on left and right side were 17.09 cm3±3.89, 17.19 cm3 ±4 respectively whereas in female were 15.64 cm3±3.5 and 15.21cm3±3.2 respectively as shown in Table 1. This shows the volume of male was significantly larger than female with P- Value = 0.012 (<0.05). Similarly, the thickness of Anterior Wall (AW) of maxillary sinus was also measured in this study and the mean value of left and right side in male were 0.16cm± 0.04 and 0.15cm± 0.03 respectively and in female were 0.12cm± 0.04 and 0.14cm± 0.02 respectively.Conclusion: This study showed that CT is a reliable method for the measurement of different dimensions of the maxillary sinus. The result showed greater mean value of volume in male than female with significant differences. So this study concluded that the measurement of volume of maxillary sinus can help in the identification of gender which can be very useful for forensic sciences.

Determining the Accuracy of the Mandibular Canal Region in 3D Biomodels Fabricated from CBCT Scanned Data: A Cadaveric Study

2019 ◽  
Vol 15 (7) ◽  
pp. 645-653 ◽  
Author(s):  
Syed Yusoff Alzawawi Syed Abd Fattah ◽  
Firdaus Hariri ◽  
Phrabhakaran Nambiar ◽  
Zulkiflee Abu Bakar ◽  
Zainal Ariff Abdul Rahman
Keyword(s):  
Computed Tomography ◽  
Gold Standard ◽  
Mandibular Canal ◽  
Mean Value ◽  
Cone Beam ◽  
Cross Sectional ◽  
Measurement Protocol ◽  
Significant Difference ◽  
The Mean ◽  
Using Data

Objective:To validate the accuracy of the mandibular canal region in 3D biomodel produced by using data obtained from Cone-Beam Computed Tomography (CBCT) of cadaveric mandibles.Methods:Six hemi-mandible samples were scanned using the i-CAT CBCT system. The scanned data was transferred to the OsiriX software for measurement protocol and subsequently into Mimics software to fabricate customized cutting jigs and 3D biomodels based on rapid prototyping technology. The hemi-mandibles were segmented into 5 dentoalveolar blocks using the customized jigs. Digital calliper was used to measure six distances surrounding the mandibular canal on each section. The same distances were measured on the corresponding cross-sectional OsiriX images and the 3D biomodels of each dentoalveolar block.Results:Statistically no significant difference was found when measurements from OsiriX images and 3D biomodels were compared to the “gold standard” -direct digital calliper measurement of the cadaveric dentoalveolar blocks. Moreover, the mean value difference of the various measurements between the different study components was also minimal.Conclusion:Various distances surrounding the mandibular canal from 3D biomodels produced from the CBCT scanned data was similar to that of direct digital calliper measurements of the cadaveric specimens.

Sign in / Sign up

Export Citation Format

Share Document