Impact of Photon-counting Detector Computed Tomography on a Quantitative Interstitial Lung Disease Machine Learning Model.

PurposeAccurate measurements of trabecular bone microarchitecture are required for a proper assessment of bone fragility. Photon-counting detector CT (PCD-CT) has different technical properties than conventional CT, resulting in higher resolution and thereby potentially enabling in-vivo measurement of trabecular microarchitecture. The purpose of this study was to quantify trabecular bone microarchitectural parameters with PCD-CT at varying radiation doses and compare this to µCT as gold standard. MethodBoth distal radii, distal tibiae, femoral heads, and two vertebrae were dissected from one human. All specimens were scanned ex-vivo on a PCD-CT system (slice increment 0.1 mm; pixel size 0.1042–0.127 mm) and a µCT system (isotropic voxel size 49–68.4 µm). The radiation doses of the PCD-CT scans were varied from 2.5 to 120 mGy based on the volume CT dose index (CTDIvol32). For the PCD-CT scans, contrast-to-noise ratio and trabecular sharpness were calculated and compared between radiation doses. µCT and PCD-CT scans were registered. The trabecular bone was then segmented from all PCD-CT and µCT scans and split into cubes with 6-mm edge length. For each cube, bone volume over total volume, trabecular thickness, trabecular number, and trabecular heterogeneity were calculated and compared between corresponding PCD-CT and µCT cubes. ResultsWith increasing dose, contrast-to-noise ratio and trabecular sharpness values increased for the PCD-CT images. Already at the lowest dose, high correlations between the trabecular microarchitectural parameters between µCT and PCD-CT were found (R2 = 0.55–0.95), which improved with increasing radiation dose (R2 = 0.76–0.96 at 20 mGy). ConclusionsPCD-CT can be used to quantify trabecular bone microarchitecture, with accuracy comparable to µCT and at clinically relevant radiation doses.

BackgroundAutomatic noninvasive oscillometric blood pressure (NIBP) devices measure mean arterial pressure (MAP); systolic and diastolic blood pressure (SBP, DBP) are algorithmically derived from MAP. The most invalid NIBP measure is SBP, yet stroke practitioners use it to manage blood pressure (BP) in accordance with thrombolysis guidelines. We determined agreement between SBP, DBP, and MAP measured manually and by NIBP in patients treated with alteplase.MethodsA multisite prospective observational study of NIBP and manual BP agreement was conducted in patients treated with alteplase immediately after bolus and infusion initiation using methods established in guidelines for the assessment of device agreement. Dual auscultatory stethoscopes were used by 2 investigators to ensure agreement with each manual BP variable and MAP was calculated using the standard formula for manual BP measures. Data were analyzed using Bland–Altman analyses and Lin concordance correlation coefficient.ResultsA total of 7 hospitals participated, collecting 5 sets of manual/NIBP BPs in 95 patients treated with alteplase (475 paired measures). Range in limits of agreement were SBP: −28.91 to 21.41 mmHg with Lin's concordance correlation coefficient 0.8; DBP: −21.0 to 19.0 mmHg with Lin's concordance correlation coefficient 0.69; and MAP: −27.5 to 16.5 mmHg with Lin's concordance correlation coefficient 0.7. There was no difference in device agreement by BP device manufacturer brand. Differences in SBP, DBP, and MAP between NIBP and manual sphygmomanometry failed to reach guideline recommendations requiring 80% of measures to fall within a 5 mmHg difference and 95% of measures to fall within a 10 mmHg difference.ConclusionNIBP devices produce significantly different BP measures then manual sphygmomanometry auscultated BP. Because NIBP devices rely on the MAP and do not directly measure SBP and DBP, definition of what constitutes safe MAP boundaries in patients treated with alteplase should be determined when automatic BP measurement is used in clinical practice.

Coronary plaque risk classification in images acquired with photon-counting-detector (PCD) CT was performed using a radiomics-based machine learning (ML) model. With IRB approval, 19 coronary CTA patients were scanned on a PCD-CT (NAEOTOM Alpha, Siemens Healthineers) with median CTDIvol of 8.02 mGy. Five types of images: virtual monoenergetic images (VMIs) at 50-keV, 70-keV, and 100-keV, iodine maps, and virtual non-contrast (VNC) images were reconstructed using an iterative reconstruction algorithm (QIR), a quantitative kernel (Qr40) and 0.6-mm/0.3-mm slice thickness/increment. Atherosclerotic plaques were segmented using semi-automatic software (Research Frontier, Siemens). Segmentation confirmation and risk stratification (low- vs high-risk) were performed by a board-certified cardiac radiologist. A total of 93 radiomic features were extracted from each image using PyRadiomics (v2.2.0b1). For each feature, a t-test was performed between low- and high-risk plaques (p<0.05 considered significant). Two significant and non-redundant features were input into a support vector machine (SVM). A leave-one-out cross-validation strategy was adopted and the classification accuracy was computed. Fifteen low-risk and ten high-risk plaques were identified by the radiologist. A total of 18, 32, 43, 16, and 55 out of 93 features in 50-keV, 70-keV, 100-keV, iodine map, and VNC images were statistically significant. A total of 17, 19, 22, 20, and 22 out of 25 plaques were classified correctly in 50-keV, 70-keV, 100-keV, iodine map, and VNC images, respectively. A ML model using 100-keV VMIs and VNC images derived from coronary PCD-CTA best automatically differentiated low- and high-risk coronary plaques.

The aim of this study was to evaluate if a high-resolution photon-counting detector computed tomography (PCD-CT) system with a 1024×1024 matrix reconstruction can improve the visualization of fine structures in the lungs compared with conventional high-resolution CT (HRCT). Twenty-two adult patients referred for clinical chest HRCT (mean CTDI vol, 13.58 mGy) underwent additional dose-matched PCD-CT (mean volume CT dose index, 13.37 mGy) after written informed consent. Computed tomography images were reconstructed at a slice thickness of 1.5 mm and an image increment of 1 mm with our routine HRCT reconstruction kernels (B46 and Bv49) at 512 and 1024 matrix sizes for conventional energy-integrating detector (EID) CT scans. For PCD-CT, routine B46 kernel and an additional sharp kernel (Q65, unavailable for EID) images were reconstructed at 1024 matrix size. Two thoracic radiologists compared images from EID and PCD-CT noting the highest level bronchus clearly identified in each lobe of the right lung, and rating bronchial wall conspicuity of third- and fourth-order bronchi. Lung nodules were also compared with the B46/EID/512 images using a 5-point Likert scale. Statistical analysis was performed using a Wilcoxon signed rank test with a P < 0.05 considered significant. Compared with B46/EID/512, readers detected higher-order bronchi using B46/PCD/1024 and Q65/PCD/1024 images for every lung lobe (P < 0.0015), but in only the right middle lobe for B46/EID/1024 (P = 0.007). Readers were able to better identify bronchial walls of the third- and fourth-order bronchi better using the Q65/PCD/1024 images (mean Likert scores of 1.1 and 1.5), which was significantly higher compared with B46/EID/1024 or B46/PCD/1024 images (mean difference, 0.8; P < 0.0001). The Q65/PCD/1024 images had a mean nodule score of 1 ± 1.3 for reader 1, and -0.1 (0.9) for reader 2, with one reader having improved nodule evaluation scores for both PCD kernels (P < 0.001), and the other reader not identifying any increased advantage over B46/EID/1024 (P = 1.0). High-resolution lung PCD-CT with 1024 image matrix reconstruction increased radiologists' ability to visualize higher-order bronchi and bronchial walls without compromising nodule evaluation compared with current chest CT, creating an opportunity for radiologists to better evaluate airway pathology.

Multifrequency bioelectrical impedance devices such as the InBody 770 (IB770) offer faster measurements and lower costs compared with other body composition assessments. This study validated measures from IB770 against the deuterium oxide dilution technique (D2O) and DXA and compared a four-compartment (4C) model using total body water (TBW) derived from IB770 compared with D2O. A total of 55 adult females (mean ± SD, age: 21.1 ± 2.6 years) completed IB770 and DXA scans and the D2O protocol. Lin's concordance correlation coefficients (CCCs), Bland-Altman analyses, and other equivalence tests evaluated agreement between IB770 and the criterion for measures of fat mass (FM), fat-free mass (FFM), and TBW individually and as part of 4C models. There was substantial agreement between IB770 and D2O for TBW (MD = Mean Difference) (MD = 0.34 L, CCC = 0.98) and between the IB770 and DXA for FM (MD = -0.22 kg, CCC = 0.99). IB770 overestimated FFM compared with DXA (MD = 3.15 kg, CCC = 0.91). Both 4C models had almost perfect agreement for FM (CCC = 0.99), FFM (CCC = 0.99), and body fat percentage (CCC = 0.99). IB770 is valid for assessing TBW and can be used within the context of a 4C model in females.

Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Cardiovascular Imaging Research Group, Medical Imaging Centre, Semmelweis University. Introduction The improved spatial resolution of photon-counting detector CT (PCD-CT) allows better image quality and less artefacts even in high-density structures, such as coronary artery stents. Purpose Our aim was to compare the prevalence of coronary artery in-stent restenosis using conventional CT and PCD-CT. Methods Our retrospective study included consecutive patients with coronary artery stents who underwent coronary CT angiography (CCTA) either with PCD-CT between January and December 2022 or with a 280-slice conventional CT scanner in 2019. The following categories were used: none, not severe, severe in-stent restenosis and stents with non-diagnostic image quality. Results Altogether 253 patients were included in our study (215 patients who underwent CCTA with a conventional CT and 38 patients who had PCD-CT examination). The mean age was 62.7±10.8 years and 46.8% were women. In-stent restenosis was detected significantly more frequently in case of PCD-CT examination, as compared to conventional CCTA (47.7% and 27.4%, respectively; p = 0.023). Among the 15 patients who underwent ultra-high resolution PCD-CT examination, only one (6.7%) had a non-diagnostic image quality for the assessment of in-stent restenosis, while 34 cases (15.8%) were non-diagnostic in the conventional CCTA group. Conclusions The better image quality provided by PCD-CT notably helps the evaluation of in-stent restenosis and can therefore contribute to the identification of mild in-stent restenosis. Moreover, the number of non-diagnostic CCTA scans can be decreased by using ultra-high resolution technique.

Placebo response complicates the interpretation of treatment response in both clinical practice and clinical trials in youth with attention-deficit/hyperactivity disorder (ADHD). In a pilot study comparing subjective ADHD symptom rating scales with scores obtained using the Quotient™ ADHD System (an objective computerized technology for assessment of hyper-activity, inattention, and impulsivity in ADHD), it was found that agreement between these 2 measures was not as strong as anticipated. This observation prompted us to evaluate placebo responses associated with subjective and objective assessments. Eligible study participants aged 6 to 14 years with a Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition ADHD diagnosis based on clinician interviews were randomized to 1 of 2 treatment sequence groups (placebo, low dose, and medium dose; or low dose, medium dose, and placebo) using either atomoxetine HCl or osmotic controlled-release (OROS) methylphenidate HCl as the active treatment in a 3-week, triple-blind (subject, parent, rater) trial. Subjects were exposed to placebo and different medication doses to evaluate the comparative sensitivity of objective and subjective measures in assessing changes in clinical condition. Placebo response was defined using 3 thresholds: any improvement, > 25% improvement, or > 40% improvement from baseline on Quotient™ Global Scaled Score (QGSS) or the ADHD Rating Scale (ADHD-RS) Total score from baseline to the visit when placebo was administered. Lin's concordance correlation coefficient was used to measure agreement between baseline and placebo scores for the objective and subjective assessments. Of 30 subjects with placebo and baseline scores, 80%, 47%, and 27% met the 3 response thresholds (ie, any, > 25%, or > 40% improvement, respectively) on the ADHD-RS Total score compared with 27%, 7%, and 0% on the QGSS. Lin's concordance correlation coefficient was 0.81 and 0.39 for the QGSS and the ADHD-RS Total score, respectively. Although larger trials are warranted, we tentatively conclude that using objective measures and higher response thresholds may enhance assay sensitivity in clinical trials and hence limit necessary patient enrollments to rule out type II statistical error.

To externally validate the Palacios' equation estimating the new baseline glomerular filtration rate (NB-GFR) after partial or radical-nephrectomy (PN, RN) for Renal cancer carcinoma (RCC). Our research group recently published two studies that investigated the association between renal function and cancer-specific survival in RCC. The first one included 3457 patients undergone RN or PN for a cT1-2 RCC coming from five high-volume centers; the second one considered 1767 patients undergone RN or PN for a cT1-4 RCC in a single high-volume center. From such datasets, available complete patients' data were used to calculate the predicted NB-GFR through the Palacios' equation: predicted NB-GFR = 35.03 + 0.65 ∙ preoperative GFR - 18.19 ∙ (if radical nephrectomy) - 0.25 ∙ age + 2.83 ∙ (if tumor size > 7cm) - 2.09 ∙ (if diabetes). The observed NB-GFR was calculated by the CKD-EPI equation on serum creatinine at 3-12months after surgery. Concordance between observed and predicted NB-GFR was evaluated by Lin's concordance correlation coefficient and Bland-Altman analysis. 2419 patients were included (1210, cohort #1; 1219, cohort #2). The median observed NB-GFR value in cohorts #1 and #2 was 73.0ml/min/1.73 m2 (IQR 56.1-90.1) and 64.2ml/min/1.73 m2 (IQR 49.6-83); the median predicted NB-GFR was 71.1ml/min/1.73 m2 (IQR 58-81.5) and 62.6ml/min/1.73m2 (IQR 47.9-75.9). The concordance line showed a slope of 0.80 and 0.86, and an intercept at 11.02 and 5.41ml/min/1.73 m2 in the cohort#1 and #2, respectively. The Palacio's equation moderately over-estimated and under-estimated NB-GFR, for values below and above the cut-off of 50ml/min/1.73 m2 and 35ml/min/1.73m2 in cohort#1 and #2. The Lin's concordance correlation coefficient was 0.79 (95% CI 0.77-0.81) and 0.83 (95% CI 0.82-0.85). We confirm the predictive performances of Palacios' equation, supporting its utilization in clinical practice.

Accuracy of a plastic facial cast fabricated with a custom tray in comparison with cone beam computed tomography

Lin's concordance correlation coefficient (CCC) is a very popular scaled index of agreement used in applied statistics. To obtain a confidence interval (CI) for the estimate of CCC, jackknifing was proposed and shown to perform well in simulation as well as in applications. However, a theoretical proof of the validity of the jackknife CI for the CCC has not been presented yet. In this note, we establish a sufficient condition for using the jackknife method to construct the CI for the CCC.

Developmental/epileptic encephalopathy with spike wave activation with sleep, formerly known as electrical status epilepticus in sleep, is an electrographic pattern in which the interictal epileptiform activity is augmented by transition to sleep. Recent studies demonstrate the utility of the first 100 seconds of sleep of long-term monitoring (LTM) as a scoring method for electrical status epilepticus in sleep. Our aim was to measure the reliability of the spike-wave index (SWI) of the first 100 seconds of sleep of routine EEG (rEEG) as a tool for diagnosis of developmental/epileptic encephalopathy with spike wave activation with sleep. Approximately three hundred forty LTMs were reviewed, and 25 studies from 25 unique patients had comparable rEEGs. Two neurophysiologists calculated the SWI of the first 100 seconds of spontaneous stage II non-random eye movement sleep, the first 5-minute bin of sleep, and three separate 5-minute bins throughout sleep in LTM. This was compared to the SWI of the first 100 seconds of sleep in rEEG. Agreement was analyzed using Lin's concordance correlation coefficient (CCC). Using 50% as a diagnostic cut-off, we observed moderate agreement between the SWI of the first 100 seconds of sleep of rEEG and three bin LTM (CCC = 0.94, 95% CI: 0.88-0.97). Agreement was slightly higher for the comparison to first bin LTM SWI (CCC = 0.96, 95% CI: 0.92-0.98) and first 100 seconds LTM SWI (CCC = 0.96, 95% CI: 0.92-0.98). This study demonstrates the first 100 seconds of sleep of rEEG technique as a time efficient diagnostic tool for patients with concern for developmental/epileptic encephalopathy with spike wave activation with sleep.

Commercial dairy farms face major challenges in safeguarding animal welfare and overall farm sustainability from environmental heat stressors. As climate change drives increased temperatures in many places, it is essential to predict the potential effects of heat stress on dairy cows to mitigate the adverse impact. This study aimed to develop an eXtreme Gradient Boosting (XGBoost) machine learning model to predict the daily milk production of 3,369 lactating dairy cows under different climatic conditions across 10 different commercial dairy farms in Australia. The duration of the dataset covered early 2019 to mid-2023, with seasonal variations. The model considered a total of 8 input parameters combining the physiological properties of cows as well as temporal and climate variables. In this study, the temperature-humidity index (THI) was incorporated, using a novel approach in which THI mean values of 5 accumulating days were considered. The model considered a mean daily THI ≥55 as a threshold point to identify a potential heat stress day (THI ≥60) and then combined the THI mean of 2 d before and 2 d after to incorporate the before- and after-effects of a potential heat stress day, defined as THI composite. The model was evaluated using combined farm data, regional farm data, and leave-one-farm-out validations, achieving high predictive accuracy (R2 up to 0.73; Lin's concordance correlation coefficient up to 0.84). The THI composite metric improved prediction accuracy by up to 21% compared with conventional rolling THI averages, demonstrating its value in forecasting milk yield under heat stress conditions. The model from this study offers a foundation for strategic planning using projected climate data to mitigate future heat-related impacts on dairy productivity.

Machine-learning modeL based on computed tomography body composition analysis for the estimation of resting energy expenditure: A pilot study.

To evaluate the inter-device reproducibility of retrobulbar blood flow measurements obtained by two commercially available CDI (color Doppler imaging) devices. The right eyes of 10 healthy volunteers were investigated. Four examiners, namely two ophthalmologists and two radiologists, performed CDI examination of the ophthalmic artery, central retinal artery and temporal short posterior ciliary arteries using both CDI devices: ESAOTE MYLAB™ and SIEMENS ANTARES STELLAR PLUS™. The peak systolic velocity (PSV), the end-diastolic velocity (EDV) and the resulting resistivity index (RI) were averaged for 3 cardiac cycles. To evaluate the reproducibility between both device measurements, the Lin's concordance correlation coefficient (CCC) was used. CCC can be expressed as the product of Pearson's r (the measure of precision) and C_b (the measure of accuracy). Results show that the inter-device reproducibility for CDI measurements is not acceptable since a poor degree of overall concordance (0.15<CCC<0.37) was obtained: accuracy was high (C_b>0.71) but overall precision low (0.18< Pearson's r <0.47). Ophthalmologists and radiologists obtained similar results. To evaluate the causal role of blood flow abnormalities in glaucoma, CDI analysis using different devices seems unreliable. CDI inter-device reproducibility seems unrelated to medical speciality of the examiners. However, to improve present results, the use of similar probes and standardized CDI instrument settings as well as a CDI images analysis by a single grader, might possibly improve the inter-device reproducibility when testing the retrobulbar blood flow velocity.

The aim of this study was to test reproducibility of the CT Pico method in a cadaveric model and to compare CT Pico measurements with a high-precision laser probe for optical scanning. The glenoid surface of ten dried cadaveric scapulae (with intact surface) was measured by and high-precision laser probe for optical scanning, the latter being assumed as a reference standard. Measurements were done according to the Pico technique, using a circle-shaped region of interest (ROI) that was placed on the inferior glenoid rim. Measurements obtained using the CT Pico method (three readers) and with laser were compared to assess differences between radiological assessment and the reference standard. Each observer performed two repeated measurements from each scapulae (20 for each observer). Mean differences between laser measurements and each CT reader were 18.4% (range, -4 to 61%) for reader 1, 12.4% (range, -15 to 64%) for reader 2, and 11% (range, -14 to 58%) for reader 3. Considering all the 60 measurements made by the three readers, 39 measurements out of 60 (65%) were outside the range [-5%; +5%] while 26 measurements (43%) are outside the range [-10%; +10%]. The largest differences (positive and negative) were +64 and -14 %, respectively. Intra-operator reproducibility was high in most cases (intraclass correlation coefficient (ICC) =0.93, ICC = 0.91, ICC = 0.93 and Lin's Concordance correlation coefficient (CCC) = 0.92, CCC = 0.90, CCC = 0.92 for reader 1 to 3, respectively. However, in five cases the CT Pico measurements showed absolute differences between the first and second measurements that exceeded 10%. The inter-observer variability for CT measurement of the glenoid surface using the CT Pico method was high when compared with laser, in the assessment of glenoid surface in cadaveric specimens, thus the CT Pico method is not reliable and could cause errors in the clinical management of the patient. Level of evidence Level II, Development of diagnostic criteria on consecutive patients (with universally applied reference "gold" standard).