Deep Learning Denoising of Low-Dose Computed Tomography Chest Images: A Quantitative and Qualitative Image Analysis
Azour, Lea; Hu, Yunan; Ko, Jane P; Chen, Baiyu; Knoll, Florian; Alpert, Jeffrey B; Brusca-Augello, Geraldine; Mason, Derek M; Wickstrom, Maj L; Kwon, Young Joon Fred; Babb, James; Liang, Zhengrong; Moore, William H
PURPOSE/OBJECTIVE:To assess deep learning denoised (DLD) computed tomography (CT) chest images at various low doses by both quantitative and qualitative perceptual image analysis. METHODS:Simulated noise was inserted into sinogram data from 32 chest CTs acquired at 100 mAs, generating anatomically registered images at 40, 20, 10, and 5 mAs. A DLD model was developed, with 23 scans selected for training, 5 for validation, and 4 for test.Quantitative analysis of perceptual image quality was assessed with Structural SIMilarity Index (SSIM) and Fréchet Inception Distance (FID). Four thoracic radiologists graded overall diagnostic image quality, image artifact, visibility of small structures, and lesion conspicuity. Noise-simulated and denoised image series were evaluated in comparison with one another, and in comparison with standard 100 mAs acquisition at the 4 mAs levels. Statistical tests were conducted at the 2-sided 5% significance level, with multiple comparison correction. RESULTS:At the same mAs levels, SSIM and FID between noise-simulated and reconstructed DLD images indicated that images were closer to a perfect match with increasing mAs (closer to 1 for SSIM, and 0 for FID).In comparing noise-simulated and DLD images to standard-dose 100-mAs images, DLD improved SSIM and FID. Deep learning denoising improved SSIM of 40-, 20-, 10-, and 5-mAs simulations in comparison with standard-dose 100-mAs images, with change in SSIM from 0.91 to 0.94, 0.87 to 0.93, 0.67 to 0.87, and 0.54 to 0.84, respectively. Deep learning denoising improved FID of 40-, 20-, 10-, and 5-mAs simulations in comparison with standard-dose 100-mAs images, with change in FID from 20 to 13, 46 to 21, 104 to 41, and 148 to 69, respectively.Qualitative image analysis showed no significant difference in lesion conspicuity between DLD images at any mAs in comparison with 100-mAs images. Deep learning denoising images at 10 and 5 mAs were rated lower for overall diagnostic image quality (P < 0.001), and at 5 mAs lower for overall image artifact and visibility of small structures (P = 0.002), in comparison with 100 mAs. CONCLUSIONS:Deep learning denoising resulted in quantitative improvements in image quality. Qualitative assessment demonstrated DLD images at or less than 10 mAs to be rated inferior to standard-dose images.
Replicability of proton MR spectroscopic imaging findings in mild traumatic brain injury: Implications for clinical applications
Chen, Anna M; Gerhalter, Teresa; Dehkharghani, Seena; Peralta, Rosemary; Gajdošík, Mia; Gajdošík, Martin; Tordjman, Mickael; Zabludovsky, Julia; Sheriff, Sulaiman; Ahn, Sinyeob; Babb, James S; Bushnik, Tamara; Zarate, Alejandro; Silver, Jonathan M; Im, Brian S; Wall, Stephen P; Madelin, Guillaume; Kirov, Ivan I
PURPOSE/OBJECTIVE:H MRS) offers biomarkers of metabolic damage after mild traumatic brain injury (mTBI), but a lack of replicability studies hampers clinical translation. In a conceptual replication study design, the results reported in four previous publications were used as the hypotheses (H1-H7), specifically: abnormalities in patients are diffuse (H1), confined to white matter (WM) (H2), comprise low N-acetyl-aspartate (NAA) levels and normal choline (Cho), creatine (Cr) and myo-inositol (mI) (H3), and correlate with clinical outcome (H4); additionally, a lack of findings in regional subcortical WM (H5) and deep gray matter (GM) structures (H6), except for higher mI in patients' putamen (H7). METHODS:26 mTBI patients (20 female, age 36.5 ± 12.5 [mean ± standard deviation] years), within two months from injury and 21 age-, sex-, and education-matched healthy controls were scanned at 3 Tesla with 3D echo-planar spectroscopic imaging. To test H1-H3, global analysis using linear regression was used to obtain metabolite levels of GM and WM in each brain lobe. For H4, patients were stratified into non-recovered and recovered subgroups using the Glasgow Outcome Scale Extended. To test H5-H7, regional analysis using spectral averaging estimated metabolite levels in four GM and six WM structures segmented from T1-weighted MRI. The Mann-Whitney U test and weighted least squares analysis of covariance were used to examine mean group differences in metabolite levels between all patients and all controls (H1-H3, H5-H7), and between recovered and non-recovered patients and their respectively matched controls (H4). Replicability was defined as the support or failure to support the null hypotheses in accordance with the content of H1-H7, and was further evaluated using percent differences, coefficients of variation, and effect size (Cohen's d). RESULTS:Patients' occipital lobe WM Cho and Cr levels were 6.0% and 4.6% higher than controls', respectively (Cho, d = 0.37, p = 0.04; Cr, d = 0.63, p = 0.03). The same findings, i.e., higher patients' occipital lobe WM Cho and Cr (both p = 0.01), but with larger percent differences (Cho, 8.6%; Cr, 6.3%) and effect sizes (Cho, d = 0.52; Cr, d = 0.88) were found in the comparison of non-recovered patients to their matched controls. For the lobar WM Cho and Cr comparisons without statistical significance (frontal, parietal, temporal), unidirectional effect sizes were observed (Cho, d = 0.07 - 0.37; Cr, d = 0.27 - 0.63). No differences were found in any metabolite in any lobe in the comparison between recovered patients and their matched controls. In the regional analyses, no differences in metabolite levels were found in any GM or WM region, but all WM regions (posterior, frontal, corona radiata, and the genu, body, and splenium of the corpus callosum) exhibited unidirectional effect sizes for Cho and Cr (Cho, d = 0.03 - 0.34; Cr, d = 0.16 - 0.51). CONCLUSIONS:H MRS biomarkers for mTBI may best be achieved by using high signal-to-noise-ratio single-voxels placed anywhere within WM. The biochemical signature of the injury, however, may differ and therefore absolute levels, rather than ratios may be preferred. Future replication efforts should further test the generalizability of these findings.
Cardiac Phase and Flow Compensation Effects on REnal Flow and Microstructure AnisotroPy MRI in Healthy Human Kidney
Sigmund, Eric E; Mikheev, Artem; Brinkmann, Inge M; Gilani, Nima; Babb, James S; Basukala, Dibash; Benkert, Thomas; Veraart, Jelle; Chandarana, Hersh
BACKGROUND:Renal diffusion-weighted imaging (DWI) involves microstructure and microcirculation, quantified with diffusion tensor imaging (DTI), intravoxel incoherent motion (IVIM), and hybrid models. A better understanding of their contrast may increase specificity. PURPOSE/OBJECTIVE:To measure modulation of DWI with cardiac phase and flow-compensated (FC) diffusion gradient waveforms. STUDY TYPE/METHODS:Prospective. POPULATION/METHODS:Six healthy volunteers (ages: 22-48 years, five females), water phantom. FIELD STRENGTH/SEQUENCE/UNASSIGNED:3-T, prototype DWI sequence with 2D echo-planar imaging, and bipolar (BP) or FC gradients. 2D Half-Fourier Single-shot Turbo-spin-Echo (HASTE). Multiple-phase 2D spoiled gradient-echo phase contrast (PC) MRI. ASSESSMENT/RESULTS:), for each tissue (cortex/medulla, segmented on b0/FA respectively), phase, and waveform (BP, FC). Monte Carlo water diffusion simulations aided data interpretation. STATISTICAL TESTS/METHODS:Mixed model regression probed differences between tissue types and pulse sequences. Univariate general linear model analysis probed variations among cardiac phases. Spearman correlations were measured between diffusion metrics and renal artery velocities. Statistical significance level was set at P < 0.05. RESULTS:, MD for FC. FA correlated significantly with velocity. Monte Carlo simulations indicated medullary measurements were consistent with a 34 μm tubule diameter. DATA CONCLUSION/CONCLUSIONS:Cardiac gating and flow compensation modulate of measurements of renal diffusion. EVIDENCE LEVEL/METHODS:2 TECHNICAL EFFICACY STAGE: 2.
FDG-PET/MRI for the preoperative diagnosis and staging of peritoneal carcinomatosis: a prospective multireader pilot study
Vietti Violi, Naik; Gavane, Somali; Argiriadi, Pamela; Law, Amy; Heiba, Sherif; Bekhor, Eliahu Y; Babb, James S; Ghesani, Munir; Labow, Daniel M; Taouli, Bachir
PURPOSE/OBJECTIVE:To assess the diagnostic performance of FDG-PET/MRI for the preoperative diagnosis and staging of peritoneal carcinomatosis (PC) using surgical Sugarbaker's PC index (PCI) as the reference in a multireader pilot study. METHODS:Fourteen adult patients (M/F: 3/11, mean age: 57â€‰Â±â€‰12Â year) with PC were prospectively included in this single-center study. Patients underwent FDG-PET/MRI prior to surgery (mean delay: 14Â d, range: 1-63Â d). Images were reviewed independently by 2 abdominal radiologists and 2 nuclear medicine physicians. The radiologists assessed contrast-enhanced abdominal MR images, while the nuclear medicine physicians assessed PET images fused with T2-weighted images. The abdomen was divided in 13 regions, scored from 0 to 3. A hybrid FDG-PET/MRI radiological PCI was created by combining the study data. Radiological PCI was compared to the surgical PCI on a per-patient and per-region basis. Inter-reader agreement was evaluated. RESULTS:Mean surgical PCI was 10â€‰Â±â€‰8 (range: 0-24). Inter-reader agreement was almost perfect for all sets for radiologic PCI (Kappa: 0.81-0.98). PCI scores for all reading sets significantly correlated with the surgical PCI score (r range: 0.57-0.74, p range:â€‰<â€‰0.001-0.003). Pooled per-patient sensitivity, specificity, and accuracy were 75%/50%/71.4% for MRI, 66.7%/50%/64.3% for FDG-PET, and 91.7%/50%/85.7% for FDG-PET/MRI, without significant difference (p value range 0.13-1). FDG-PET/MRI achieved 100% sensitivity and 100% specificity for a cutoff PCI of 20. Per-region sensitivity and accuracy were lower: 37%/61.8% for MRI, 17.8%/64.3% for FDG-PET, and 52.7%/60.4% for FDG-PET/MRI, with significantly higher sensitivity for FDG-PET/MRI. Per-region specificity was higher for FDG-PET (95%) compared to MRI (78.4%) and FDG-PET/MRI (66.5%). CONCLUSION/CONCLUSIONS:FDG-PET/MRI achieved an excellent diagnostic accuracy per-patient and weaker performance per-region for detection of PC. The added value of PET/MRI compared to MRI and FDG-PET remains to be determined.
Low-field 0.55Â T MRI for assessment of pulmonary groundglass and fibrosis-like opacities: Inter-reader and inter-modality concordance
Azour, Lea; Condos, Rany; Keerthivasan, Mahesh B; Bruno, Mary; Pandit Sood, Terlika; Landini, Nicholas; Silverglate, Quinn; Babb, James; Chandarana, Hersh; Moore, William H
PURPOSE/OBJECTIVE:To evaluate detection and characterization of groundglass and fibrosis-like opacities imaged by non-contrast 0.55 Tesla MRI, and versus clinically-acquired chest CT images, in a cohort of post-Covid patients. MATERIALS AND METHODS/METHODS:64 individuals (26 women, mean age 53Â Â±Â 14Â years, range 19-85) with history of Covid-19 pneumonia were recruited through a survivorship registry, with 106 non-contrast low-field 0.55Â T cardiopulmonary MRI exams acquired from 9/8/2020-9/28/2021. MRI exams were obtained at an average interval of 9.5Â Â±Â 4.5Â months from initial symptom report (range 1-18Â months). Of these, 20 participants with 22 MRI exams had corresponding clinically-acquired CT chest imaging obtained within 30Â days of MRI (average interval 18Â Â±Â 9Â days, range 0-30). MR and CT images were reviewed and scored by two thoracic radiologists, for presence and extent of lung opacity by quadrant, opacity distribution, and presence versus absence of fibrosis-like subpleural reticulation and subpleural lines. Scoring was performed for each of four lung quadrants: right upper and middle lobe, right lower lobe, left upper lobe and lingula, and left lower lobe. Agreement between readers and modalities was assessed with simple and linear weighted Cohen's kappa (k) coefficients. RESULTS:Inter-reader concordance on CT for opacity presence, opacity extent, opacity distribution, and presence of subpleural lines and reticulation was 99%, 78%, 97%, 99%, and 94% (k 0.96, 0.86, 0.94, 0.97, 0.89), respectively. Inter-reader concordance on MR, among all 106 exams, for opacity presence, opacity extent, opacity distribution, and presence of subpleural lines and reticulation was 85%, 48%, 70%, 86%, and 76% (k 0.57, 0.32, 0.46, 0.47, 0.37), respectively. Inter-modality agreement between CT and MRI for opacity presence, opacity extent, opacity distribution, and presence subpleural lines and reticulation was 86%, 52%, 79%, 93%, and 76% (k 0.43, 0.63, 0.65, 0.80, 0.52). CONCLUSION/CONCLUSIONS:Low-field 0.55Â T non-contrast MRI demonstrates fair to moderate inter-reader concordance, and moderate to substantial inter-modality agreement with CT, for detection and characterization of groundglass and fibrosis-like opacities.
Association between lower body temperature and increased tau pathology in cognitively normal older adults
Blessing, Esther M; Parekh, Ankit; Betensky, Rebecca A; Babb, James; Saba, Natalie; Debure, Ludovic; Varga, Andrew W; Ayappa, Indu; Rapoport, David M; Butler, Tracy A; de Leon, Mony J; Wisniewski, Thomas; Lopresti, Brian J; Osorio, Ricardo S
BACKGROUND:Preclinical studies suggest body temperature (Tb) and consequently brain temperature has the potential to bidirectionally interact with tau pathology in Alzheimer's Disease (AD). Tau phosphorylation is substantially increased by small (<1â€¯Â°C) decreases in temperature within the human physiological range, and thermoregulatory nuclei are affected by tau pathology early in the AD continuum. In this study we evaluated whether Tb (as a proxy for brain temperature) is cross-sectionally associated with clinically utilized markers of tau pathology in cognitively normal older adults. METHODS:Tb was continuously measured with ingestible telemetry sensors for 48-h. This period also included two nights of nocturnal polysomnography to delineate whether Tb during waking vs sleep is differentially associated with tau pathology. Tau phosphorylation was assessed with plasma and cerebrospinal fluid (CSF) tau phosphorylated at threonine 181 (P-tau), sampled the day following Tb measurement. In addition, neurofibrillary tangle (NFT) burden in early Braak stage regions was imaged with PET-MR using the [18F]MK-6240 radiotracer on average one month later. RESULTS:Lower Tb was associated with increased NFT burden, as well as increased plasma and CSF P-tau levels (pâ€¯<â€¯0.05). NFT burden was associated with lower Tb during waking (pâ€¯<â€¯0.05) but not during sleep intervals. Plasma and CSF Ptau levels were highly correlated with each other (pâ€¯<â€¯0.05), and both variables were correlated with tau tangle radiotracer uptake (pâ€¯<â€¯0.05). CONCLUSIONS:These results, the first available for human, suggest that lower Tb in older adults may be associated with increased soluble and aggregated tau pathology. Our findings add to the substantial preclinical literature associating lower body and brain temperature with tau hyperphosphorylation. CLINICAL TRIAL NUMBER/BACKGROUND:NCT03053908.
Outcomes of Incidental Lung Nodules With Structured Recommendations and Electronic Tracking
Bagga, Barun; Fansiwala, Kush; Thomas, Shailin; Chung, Ryan; Moore, William H; Babb, James S; Horwitz, Leora I; Blecker, Saul; Kang, Stella K
OBJECTIVE:To evaluate the impact of structured recommendations on follow-up completion for incidental lung nodules (ILNs). METHODS:Patients with ILNs before and after implementation of structured Fleischner recommendations and electronic tracking were sampled randomly. The cohorts were compared for imaging follow-up. Multivariable logistic regression was used to assess appropriate follow-up and loss to follow-up, with independent variables including use of structured recommendations or tracking, age, gender, race, ethnicity, setting of the index test (inpatient, outpatient, emergency department), smoking history, and nodule features. RESULTS:In all, 1,301 patients met final inclusion criteria, including 255 patients before and 1,046 patients after structured recommendations or tracking. Baseline differences were found in the pre- and postintervention groups, with smaller ILNs and younger age after implementing structured recommendations. Comparing pre- versus postintervention outcomes, 40.0% (100 of 250) versus 29.5% (309 of 1,046) of patients had no follow-up despite Fleischner indications for imaging (PÂ = .002), and among the remaining patients, 56.6% (82 of 145) versus 75.0% (553 of 737) followed up on time (P < .001). Delayed follow-up was more frequent before intervention. Differences postintervention were mostly accounted for by nodules â‰¤ 8 mm in the outpatient setting (P < .001). In multivariable analysis, younger age, White race, outpatient setting, and larger nodule size showed significant association with appropriate follow-up completion (P < .015), but structured recommendations did not. Similar results applied for loss to follow-up. DISCUSSION/CONCLUSIONS:Consistent use of structured reporting is likely key to mitigate selection bias when benchmarking rates of appropriate follow-up of ILN. Emergency department patients and inpatients are at high risk of missed or delayed follow-up despite structured recommendations.
Inter-Reader Variability of Volumetric Subsolid Pulmonary Nodule Radiomic Features
Azour, Lea; Moore, William H; O'Donnell, Thomas; Truong, Mylene T; Babb, James; Niu, Bowen; Wimmer, Andreas; Kiumehr, Saman; Ko, Jane P
OBJECTIVE:To evaluate the inter-observer consistency for subsolid pulmonary nodule radiomic features. MATERIALS AND METHODS/METHODS:Subsolid nodules were selected by reviewing radiology reports of CT examinations performed December 1, 2015 to April 1, 2016. Patients with CTs at two time points were included in this study. There were 55 patients with subsolid nodules, of whom 14 had two nodules. Of 69 subsolid nodules, 66 were persistent at the second time point, yielding 135 lesions for segmentation. Two thoracic radiologists and an imaging fellow segmented the lesions using a semi-automated volumetry algorithm (Syngo.via Vb20, Siemens). Coefficient of variation (CV) was used to assess consistency of 91 quantitative measures extracted from the subsolid nodule segmentations, including first and higher order texture features. The accuracy of segmentation was visually graded by an experienced thoracic radiologist. Influencing factors on radiomic feature consistency and segmentation accuracy were assessed using generalized estimating equation analyses and the Exact Mann-Whitney test. RESULTS:Mean patient age was 71 (38-93 years), with 39 women and 16 men. Mean nodule volume was 1.39mL, range .03-48.2mL, for 135 nodules. Several radiomic features showed high inter-reader consistency (CV<5%), including entropy, uniformity, sphericity, and spherical disproportion. Descriptors such as surface area and energy had low consistency across inter-reader segmentations (CV>10%). Nodule percent solid component and attenuation influenced inter-reader variability of some radiomic features. The presence of contrast did not significantly affect the consistency of subsolid nodule radiomic features. Near perfect segmentation, within 5% of actual nodule size, was achieved in 68% of segmentations, and very good segmentation, within 25% of actual nodule size, in 94%. Morphologic features including nodule margin and shape (each p <0.01), and presence of air bronchograms (p = 0.004), bubble lucencies (p = 0.02) and broad pleural contact (p < 0.01) significantly affected the probability of near perfect segmentation. Stroke angle (p = 0.001) and length (p < 0.001) also significantly influenced probability of near perfect segmentation. CONCLUSIONS:The inter-observer consistency of radiomic features for subsolid pulmonary nodules varies, with high consistency for several features, including sphericity, spherical disproportion, and first and higher order entropy, and normalized non-uniformity. Nodule morphology influences the consistency of subsolid nodule radiomic features, and the accuracy of subsolid nodule segmentation.
T1 and T2 quantification using magnetic resonance fingerprinting in mild traumatic brain injury
Gerhalter, Teresa; Cloos, Martijn; Chen, Anna M; Dehkharghani, Seena; Peralta, Rosemary; Babb, James S; Zarate, Alejandro; Bushnik, Tamara; Silver, Jonathan M; Im, Brian S; Wall, Stephen; Baete, Steven; Madelin, Guillaume; Kirov, Ivan I
OBJECTIVES/OBJECTIVE:To assess whether MR fingerprinting (MRF)-based relaxation properties exhibit cross-sectional and prospective correlations with patient outcome and compare the results with those from DTI. METHODS:from MRF were compared in 12Â gray and white matter regions with Mann-Whitney tests. Bivariate associations between MR measures and outcome were assessed using the Spearman correlation and logistic regression. RESULTS:, accounted for five of the six MR measures with the highest utility for identification of non-recovered patients at timepoint 2 (AUCâ€‰>â€‰0.80). CONCLUSION/CONCLUSIONS:, FA, and ADC for predicting 3-month outcome after mTBI. KEY POINTS/CONCLUSIONS:, and FA.
Factors predicting hip joint aspiration yield or "dry taps" in patients with total hip arthroplasty
Ong, Justin; Tang, Alex; Rozell, Joshua C; Babb, James S; Schwarzkopf, Ran; Lin, Dana
BACKGROUND:Image-guided joint aspirations used to assist the diagnosis of periprosthetic joint infection (PJI) may commonly result in a dry tap-or insufficient fluid for culture and cell count analysis. Dry tap aspirations are painful and invasive for patients and often utilize a subsequent saline lavage to obtain a microbiology sample. Currently, there is a paucity of the literature addressing predictors that could suggest whether a dry tap will occur. The purpose of this study was to examine the effects of various factors on "dry tap" occurrence in patients with suspected PJI following total hip arthroplasty (THA). METHODS:A retrospective review was performed among THA patients suspected for PJI who received image-guided joint aspiration procedures at our institution from May 2016 to February 2020. The procedural factors included the imaging modality used for aspiration, anatomic approach, needle gauge size used, and the presence of a trainee. The patient-specific factors included number of prior ipsilateral hip surgeries, femoral head size, ESR/CRP values, and BMI. RESULTS:In total, 336 patients met our inclusion criteria. One hundred and twenty hip aspirations resulted in a dry tap (35.7%) where the patients underwent a saline lavage. Among the procedural and patient-specific factors, none of the factors were found to be statistically different between the two cohorts nor conferred any greater odds of a dry tap occurring. CONCLUSION/CONCLUSIONS:No associations with dry tap occurrence were found among the procedural and patient-specific factors studied. Further research is needed to identify additional factors that may be more predictive of dry taps.