scholarly journals Comparing two artificial intelligence software packages for normative brain volumetry in memory clinic imaging

2022 ◽  
Author(s):  
Lara A. M. Zaki ◽  
Meike W. Vernooij ◽  
Marion Smits ◽  
Christine Tolman ◽  
Janne M. Papma ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Sensitivity And Specificity ◽  
Brain Mri ◽  
Intracranial Volume ◽  
Diagnostic Confidence ◽  
Brain Volumetry ◽  
Software Packages ◽  
Volume Output ◽  
Dementia Diagnostics ◽  
Subtype Differences

Abstract Purpose To compare two artificial intelligence software packages performing normative brain volumetry and explore whether they could differently impact dementia diagnostics in a clinical context. Methods Sixty patients (20 Alzheimer’s disease, 20 frontotemporal dementia, 20 mild cognitive impairment) and 20 controls were included retrospectively. One MRI per subject was processed by software packages from two proprietary manufacturers, producing two quantitative reports per subject. Two neuroradiologists assigned forced-choice diagnoses using only the normative volumetry data in these reports. They classified the volumetric profile as “normal,” or “abnormal”, and if “abnormal,” they specified the most likely dementia subtype. Differences between the packages’ clinical impact were assessed by comparing (1) agreement between diagnoses based on software output; (2) diagnostic accuracy, sensitivity, and specificity; and (3) diagnostic confidence. Quantitative outputs were also compared to provide context to any diagnostic differences. Results Diagnostic agreement between packages was moderate, for distinguishing normal and abnormal volumetry (K = .41–.43) and for specific diagnoses (K = .36–.38). However, each package yielded high inter-observer agreement when distinguishing normal and abnormal profiles (K = .73–.82). Accuracy, sensitivity, and specificity were not different between packages. Diagnostic confidence was different between packages for one rater. Whole brain intracranial volume output differed between software packages (10.73%, p < .001), and normative regional data interpreted for diagnosis correlated weakly to moderately (rs = .12–.80). Conclusion Different artificial intelligence software packages for quantitative normative assessment of brain MRI can produce distinct effects at the level of clinical interpretation. Clinics should not assume that different packages are interchangeable, thus recommending internal evaluation of packages before adoption.

scholarly journals Detection of Microcalcifications in Spiral Breast Computed Tomography with Photon-Counting Detector Is Feasible: A Specimen Study

Diagnostics ◽  
2021 ◽  
Vol 11 (5) ◽  
pp. 848
Author(s):  
Matthias Wetzl ◽  
Evelyn Wenkel ◽  
Eva Balbach ◽  
Ebba Dethlefsen ◽  
Arndt Hartmann ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Sensitivity And Specificity ◽  
Breast Biopsy ◽  
Photon Counting ◽  
Diagnostic Confidence ◽  
Voxel Size ◽  
Photon Counting Detector ◽  
Smooth Kernel ◽  
Higher Sensitivity ◽  
Breast Computed Tomography

The primary objective of the study was to compare a spiral breast computed tomography system (SBCT) to digital breast tomosynthesis (DBT) for the detection of microcalcifications (MCs) in breast specimens. The secondary objective was to compare various reconstruction modes in SBCT. In total, 54 breast biopsy specimens were examined with mammography as a standard reference, with DBT, and with a dedicated SBCT containing a photon-counting detector. Three different reconstruction modes were applied for SBCT datasets (Recon1 = voxel size (0.15 mm)3, smooth kernel; Recon2 = voxel size (0.05 mm)3, smooth kernel; Recon3 = voxel size (0.05 mm)3, sharp kernel). Sensitivity and specificity of DBT and SBCT for the detection of suspicious MCs were analyzed, and the McNemar test was used for comparisons. Diagnostic confidence of the two readers (Likert Scale 1 = not confident; 5 = completely confident) was analyzed with ANOVA. Regarding detection of MCs, reader 1 had a higher sensitivity for DBT (94.3%) and Recon2 (94.9%) compared to Recon1 (88.5%; p < 0.05), while sensitivity for Recon3 was 92.4%. Respectively, reader 2 had a higher sensitivity for DBT (93.0%), Recon2 (92.4%), and Recon3 (93.0%) compared to Recon1 (86.0%; p < 0.05). Specificities ranged from 84.7–94.9% for both readers (p > 0.05). The diagnostic confidence of reader 1 was better with SBCT than with DBT (DBT 4.48 ± 0.88, Recon1 4.77 ± 0.66, Recon2 4.89 ± 0.44, and Recon3 4.75 ± 0.72; DBT vs. Recon1/2/3: p < 0.05), while reader 2 found no differences. Sensitivity and specificity for the detection of MCs in breast specimens is equal for DBT and SBCT when a small voxel size of (0.05 mm)3 is used with an equal or better diagnostic confidence for SBCT compared to DBT.

Abstract MP100: Arterial Stiffness and Pressure Amplification Associated with Markers of Cerebral Microvascular Disease: The Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS)

Circulation ◽  
2016 ◽  
Vol 133 (suppl_1) ◽  
Author(s):  
Priya Palta ◽  
Jingkai Wei ◽  
Michelle Meyer ◽  
Melinda C Power ◽  
Jennifer A Deal ◽  
...  
Keyword(s):  
Older Adults ◽  
Arterial Stiffness ◽  
Pulse Pressure ◽  
Aortic Stiffness ◽  
Brain Mri ◽  
Intracranial Volume ◽  
Sampling Weights ◽  
Arterial Stiffening ◽  
Lacunar Infarcts ◽  
Pressure Amplification

Introduction: Small vessel disease is associated with decreased cognitive function, possibly differential by race. Age-related central arterial stiffening increases pulsatility resulting in hypoperfusion, microvascular damage and remodeling in the brain, potentially impairing cognition. We examined if arterial stiffness and pressure amplification are associated with lacunar infarcts and greater volumes of white matter hyperintensities (WMH) in a sample of Caucasian and African American (AA) older adults. Methods: We analyzed a cross-sectional sample of ARIC participants aged 67-90 years (n=1486) from visit 5 (2011-2013), with brain magnetic resonance imaging (MRI). The Omron VP-1000 Plus was used to measure aortic stiffness (carotid-femoral pulse wave velocity [cfPWV]) and pressure amplification measures (pulse pressure amplification [PPA], central pulse pressure [cPP], and estimated central systolic blood pressure [cSBP]). Aortic stiffness and pressure amplification were dichotomized at race-specific 25th percentile cut points. Brain MRI using 3D-1.5T equipment quantified the presence of lacunar infarcts and volumes of WMH following a standardized protocol. Logistic regression, adjusted for age, sex, education, ApoE4, heart rate, smoking and body mass index, was used to quantify the odds of lacunar infarcts in participants with high vs. low cfPWV, cPP, cSBP, and low vs. high PPA. Linear regression models, additionally adjusted for intracranial volume, estimated the difference in log-transformed volumes of WMH among participants with high vs. low cfPWV, cPP, cSBP, and low vs. high PPA. Probability sampling weights for an MRI were included to allow for generalizability to the full visit 5 cohort. Results: Among the 1486 participants with a brain MRI (mean age: 76, 41% male, 26% AA), measures of aortic stiffness and pressure amplification were associated with lacunar infarcts in Caucasians, but not in AAs. Caucasian participants with a high cfPWV had greater odds of lacunar infarcts (Odds Ratio [OR] =2.02, 95% confidence interval [CI]: 1.23, 2.20). Caucasians with high cSBP had higher odds of lacunar infarcts (OR=1.72, 95% CI: 1.10, 2.69). In Caucasians, high cfPWV was associated with a 21% (95% CI: 6, 38) greater volume of WMH as compared to a low cfPWV; high cSBP was associated with a 28% (95% CI: 14, 45) greater volume of WMH compared to a low cSBP. In AAs, high cfPWV was associated with a 32% (95% CI: 7, 62) greater volume of WMH as compared to low cfPWV. Cerebral microvascular imaging markers did not differ quantitatively with measures of PPA and cPP. Conclusions: Central arterial stiffening and pressure amplification are plausible microvascular contributors to cognitive aging, providing new information on modifiable pathways for previously observed associations between cardiovascular disease risk factors and the rates of cognitive decline and dementia among older adults.

Fish Intake and MRI Burden of Cerebrovascular Disease in Older Adults

Neurology ◽  
2021 ◽  
pp. 10.1212/WNL.0000000000012916
Author(s):  
Aline Thomas ◽  
Fabrice Crivello ◽  
Bernard Mazoyer ◽  
Stephanie Debette ◽  
Christophe Tzourio ◽  
...  
Keyword(s):  
Factor Analysis ◽  
Cerebrovascular Disease ◽  
Brain Mri ◽  
Large Population ◽  
Population Based ◽  
Fish Intake ◽  
Mixed Data ◽  
Perivascular Spaces ◽  
Intracranial Volume ◽  
Cross Sectional

Background and Objective:Fish intake may prevent cerebrovascular disease (CVD), yet the mechanisms are unclear, especially regarding its impact on subclinical damage. Assuming that fish may have pleiotropic effect on cerebrovascular health, we investigated the association of fish intake with global CVD burden based on brain MRI markers.Methods:This cross-sectional analysis included participants from the Three-City Dijon population-based cohort (aged ≥65 years) without dementia, stroke, or history of hospitalized cardiovascular disease, who underwent brain MRI with automated assessment of white matter hyperintensities, visual detection of covert infarcts, and grading of dilated perivascular spaces. Fish intake was assessed through a frequency questionnaire and the primary outcome measure was defined as the first component of a factor analysis of mixed data applied to MRI markers. The association of fish intake with the CVD burden indicator was studied using linear regressions.Results:In total, 1,623 participants (mean age, 72.3 years; 63% women) were included. The first component of factor analysis (32.4% of explained variance) was associated with higher levels of all three MRI markers. Higher fish intake was associated with lower CVD burden. In a model adjusted for total intracranial volume, compared to participants consuming fish <1 per week, those consuming fish 2-3 and ≥4 times per week had a β = -0.19 (95% CI, -0.37; -0.01) and β = -0.30 (-0.57; -0.03) lower indicator of CVD burden, respectively (P trend <0.001). We found evidence of effect modification by age, so that the association of fish to CVD was stronger in younger participants (65-69 years) and not significant in participants aged ≥75 years. For comparison, in the younger age group, consuming fish 2-3 times a week was roughly equivalent (in opposite direction) to the effect of hypertension.Discussion:In this large population-based study, higher frequency of fish intake was associated with lower CVD burden, especially among participants younger than 75 years, suggesting a beneficial effect on brain vascular health before manifestation of overt brain disease.Classification of Evidence:This study provides Class II evidence that in individuals without stroke or dementia, higher fish intake is associated with lower subclinical CVD at MRI.

1267 Utility of Artificial Intelligence in the Cystoscopic Detection of Bladder Cancer: A Systematic Review and Meta-Analysis

2021 ◽  
Vol 108 (Supplement_6) ◽  
Author(s):  
S Ganesananthan ◽  
S Ganesananthan ◽  
B S Simpson ◽  
J M Norris
Keyword(s):  
Artificial Intelligence ◽  
Bladder Cancer ◽  
Sensitivity And Specificity ◽  
Meta Analysis ◽  
External Validation ◽  
High Sensitivity ◽  
Diagnostic Odds Ratio ◽  
Pooled Analysis ◽  
Likelihood Ratios ◽  
Novel Technology

Abstract Aim Detection of suspected bladder cancer at diagnostic cystoscopy is challenging and is dependent on clinician skill. Artificial Intelligence (AI) algorithms, specifically, machine learning and deep learning, have shown promise in accurate classification of pathological images in various specialties. However, utility of AI for urothelial cancer diagnosis is unknown. Here, we aimed to systematically review the extant literature in this field and quantitively summarise the role of these algorithms in bladder cancer detection. Method The EMBASE, PubMed and CENTRAL databases were searched up to December 22nd 2020 , in accordance with the PRISMA guidelines, for studies that evaluated AI algorithms for cystoscopic diagnosis of bladder cancer. Random-effects meta-analysis was performed to summarise eligible studies. Risk of Bias was assessed using the QUADAS-2 tool. Results Five from 6715 studies met criteria for inclusion. Pooled sensitivity and specificity values were 0.93 (95% CI 0.89–0.95) and 0.93 (95% CI 0.80–0.89) respectively. Pooled positive likelihood and negative likelihood ratios were 14 (95% CI 4.3–44) and 0.08 (95% CI: 0.05–0.11), respectively. Pooled diagnostic odds ratio was 182 (95% CI 61–546). Summary AUC curve value was 0.95 (95% CI 0.93–0.97). No significant publication bias was noted. Conclusions In summary, AI algorithms performed very well in detection of bladder cancer in this pooled analysis, with high sensitivity and specificity values. However, as with other clinical AI usage, further external validation through deployment in real clinical situations is essential to assess true applicability of this novel technology.

The Potential Role of Peritumoral Apparent Diffusion Coefficient Evaluation in Differentiating Glioblastoma and Solitary Metastatic Lesions of the Brain

2021 ◽  
Vol 17 ◽  
Author(s):  
Murat Tepe ◽  
Suzan Saylisoy ◽  
Ugur Toprak ◽  
Ibrahim Inan
Keyword(s):  
Diffusion Coefficient ◽  
White Matter ◽  
Apparent Diffusion Coefficient ◽  
Sensitivity And Specificity ◽  
Brain Mri ◽  
Solitary Metastasis ◽  
Metastatic Lesions ◽  
Institutional Review ◽  
Apparent Diffusion ◽  
Adc Values

Objective: Differentiating glioblastoma (GBM) and solitary metastasis is not always possible using conventional magnetic resonance imaging (MRI) techniques. In conventional brain MRI, GBM and brain metastases are lesions with mostly similar imaging findings. In this study, we investigated whether apparent diffusion coefficient (ADC) ratios, ADC gradients, and minimum ADC values in the peritumoral edema tissue can be used to discriminate between these two tumors. Methods: This retrospective study was approved by the local institutional review board with a waiver of written informed consent. Prior to surgical and medical treatment, conventional brain MRI and diffusion-weighted MRI (b = 0 and b = 1000) images were taken from 43 patients (12 GBM and 31 solitary metastasis cases). Quantitative ADC measurements were performed on the peritumoral tissue from the nearest segment to the tumor (ADC1), the middle segment (ADC2), and the most distant segment (ADC3). The ratios of these three values were determined proportionally to calculate the peritumoral ADC ratios. In addition, these three values were subtracted from each other to obtain the peritumoral ADC gradients. Lastly, the minimum peritumoral and tumoral ADC values, and the quantitative ADC values from the normal appearing ipsilateral white matter, contralateral white matter and ADC values from cerebrospinal fluid (CSF) were recorded. Results: For the differentiation of GBM and solitary metastasis, ADC3 / ADC1 was the most powerful parameter with a sensitivity of 91.7% and specificity of 87.1% at the cut-off value of 1.105 (p < 0.001), followed by ADC3 / ADC2 with a cut-off value of 1.025 (p = 0.001), sensitivity of 91.7%, and specificity of 74.2%. The cut-off, sensitivity and specificity of ADC2 / ADC1 were 1.055 (p = 0.002), 83.3%, and 67.7%, respectively. For ADC3 – ADC1, the cut-off value, sensitivity and specificity were calculated as 150 (p < 0.001), 91.7% and 83.9%, respectively. ADC3 – ADC2 had a cut-off value of 55 (p = 0.001), sensitivity of 91.7%, and specificity of 77.4 whereas ADC2 – ADC1 had a cut-off value of 75 (p = 0.003), sensitivity of 91.7%, and specificity of 61.3%. Among the remaining parameters, only the ADC3 value successfully differentiated between GBM and metastasis (GBM 1802.50 ± 189.74 vs. metastasis 1634.52 ± 212.65, p = 0.022). Conclusion: The integration of the evaluation of peritumoral ADC ratio and ADC gradient into conventional MR imaging may provide valuable information for differentiating GBM from solitary metastatic lesions.

scholarly journals Simple, Mobile-based Artificial Intelligence Algorithm in the detection of Diabetic Retinopathy (SMART) study

2020 ◽  
Vol 8 (1) ◽  
pp. e000892 ◽  
Author(s):  
Bhavana Sosale ◽  
Sosale Ramachandra Aravind ◽  
Hemanth Murthy ◽  
Srikanth Narayana ◽  
Usha Sharma ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Diabetic Retinopathy ◽  
Sensitivity And Specificity ◽  
Smart Phone ◽  
High Sensitivity ◽  
Ground Truth ◽  
Cross Sectional Study ◽  
Retinal Images ◽  
Cross Sectional ◽  
Smart Study

IntroductionThe aim of this study is to evaluate the performance of the offline smart phone-based Medios artificial intelligence (AI) algorithm in the diagnosis of diabetic retinopathy (DR) using non-mydriatic (NM) retinal images.MethodsThis cross-sectional study prospectively enrolled 922 individuals with diabetes mellitus. NM retinal images (disc and macula centered) from each eye were captured using the Remidio NM fundus-on-phone (FOP) camera. The images were run offline and the diagnosis of the AI was recorded (DR present or absent). The diagnosis of the AI was compared with the image diagnosis of five retina specialists (majority diagnosis considered as ground truth).ResultsAnalysis included images from 900 individuals (252 had DR). For any DR, the sensitivity and specificity of the AI algorithm was found to be 83.3% (95% CI 80.9% to 85.7%) and 95.5% (95% CI 94.1% to 96.8%). The sensitivity and specificity of the AI algorithm in detecting referable DR (RDR) was 93% (95% CI 91.3% to 94.7%) and 92.5% (95% CI 90.8% to 94.2%).ConclusionThe Medios AI has a high sensitivity and specificity in the detection of RDR using NM retinal images.

scholarly journals A bibliometric and visual analysis of artificial intelligence technologies-enhanced brain MRI research

2020 ◽  
Author(s):  
Xieling Chen ◽  
Xinxin Zhang ◽  
Haoran Xie ◽  
Xiaohui Tao ◽  
Fu Lee Wang ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Visual Analysis ◽  
Brain Mri

scholarly journals Volume and shape of subcortical grey matter structures related to headache: A cross-sectional population-based imaging study in the Nord-Trøndelag Health Study

Cephalalgia ◽  
2018 ◽  
Vol 39 (2) ◽  
pp. 173-184 ◽  
Author(s):  
Andreas Kattem Husøy ◽  
Carl Pintzka ◽  
Live Eikenes ◽  
Asta K Håberg ◽  
Knut Hagen ◽  
...  
Keyword(s):  
Brain Mri ◽  
Depression Scale ◽  
Imaging Study ◽  
Population Based ◽  
Tension Type Headache ◽  
Analysis Of Covariance ◽  
Health Study ◽  
Intracranial Volume ◽  
Cross Sectional ◽  
Subcortical Nuclei

Background The relationship between subcortical nuclei and headache is unclear. Most previous studies were conducted in small clinical migraine samples. In the present population-based MRI study, we hypothesized that headache sufferers exhibit reduced volume and deformation of the nucleus accumbens compared to non-sufferers. In addition, volume and deformation of the amygdala, caudate, hippocampus, pallidum, putamen and thalamus were examined. Methods In all, 1006 participants (50–66 years) from the third Nord-Trøndelag Health Survey, were randomly selected to undergo a brain MRI at 1.5 T. Volume and shape of the subcortical nuclei from T1 weighted 3D scans were obtained in FreeSurfer and FSL. The association with questionnaire-based headache categories (migraine and tension-type headache included) was evaluated using analysis of covariance. Individuals not suffering from headache were used as controls. Age, sex, intracranial volume and Hospital Anxiety and Depression Scale were used as covariates. Results No effect of headache status on accumbens volume and shape was present. Exploratory analyses showed significant but small differences in volume of caudate and putamen and in putamen shape between those with non-migrainous headache and the controls. A post hoc analysis showed that caudate volume was strongly associated with white matter hyperintensities. Conclusion We did not confirm our hypothesis that headache sufferers have smaller volume and different shape of the accumbens compared to non-sufferers. No or only small differences in volume and shape of subcortical nuclei between headache sufferers and non-sufferers appear to exist in the general population.

scholarly journals Artificial Intelligence System Approaching Neuroradiologist-level Differential Diagnosis Accuracy at Brain MRI

Radiology ◽  
2020 ◽  
Vol 295 (3) ◽  
pp. 626-637 ◽  
Author(s):  
Andreas M. Rauschecker ◽  
Jeffrey D. Rudie ◽  
Long Xie ◽  
Jiancong Wang ◽  
Michael Tran Duong ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Differential Diagnosis ◽  
Brain Mri ◽  
Artificial Intelligence System ◽  
Intelligence System ◽  
Diagnosis Accuracy

scholarly journals Sensitivity and specificity of artificial intelligence with Microsoft Azure in detecting pneumothorax in emergency department: A pilot study

2020 ◽  
pp. 102490792094899
Author(s):  
Kwok Hung Alastair Lai ◽  
Shu Kai Ma
Keyword(s):  
Artificial Intelligence ◽  
Emergency Department ◽  
Sensitivity And Specificity ◽  
National Institutes Of Health ◽  
Dramatic Improvement ◽  
X Ray ◽  
Training Images ◽  
Normal Chest ◽  
Chest X Ray ◽  
Microsoft Azure

Background: Artificial intelligence is becoming an increasingly important tool in different medical fields. This article aims to evaluate the sensitivity and specificity of artificial intelligence trained with Microsoft Azure in detecting pneumothorax. Methods: A supervised learning artificial intelligence is trained with a collection of X-ray images of pneumothorax from National Institutes of Health chest X-ray dataset online. A subset of the image dataset focused on pneumothorax is used in training. Two artificial intelligence programs are trained with different numbers of training images. After the training, a collection of pneumothorax X-ray images from patient attending emergency department is retrieved through the Clinical Data Analysis & Reporting System. In total, 115 pneumothorax patients and 60 normal inpatients are recruited. The pneumothorax chest X-ray and the resolution chest X-ray of the above patient group and a collection of normal chest X-ray from inpatients without pneumothorax will be retrieved, and these three sets of images will then undergo testing by artificial intelligence programs to give a probability of being a pneumothorax X-ray. Results: The sensitivity of artificial intelligence-one is 33.04%, and the specificity is at least 61.74%. The sensitivity of artificial intelligence-two is 46.09%, and the specificity is at least 71.30%. The dramatic improvement of 46.09% in sensitivity and improvement of 15.48% in specificity by addition of around 1000 X-ray images is encouraging. The mean improvement of AI-two over AI-one is 19.7% increase in probability difference. Conclusions: We should not rely on artificial intelligence in diagnosing pneumothorax X-ray solely by our models and more training should be expected to explore its full function.

Sign in / Sign up

Export Citation Format

Share Document