CT Volume Research Articles

UniMiSS+: Universal Medical Self-Supervised Learning From Cross-Dimensional Unpaired Data.

Self-supervised learning (SSL) opens up huge opportunities for medical image analysis that is well known for its lack of annotations. However, aggregating massive (unlabeled) 3D medical images like computerized tomography (CT) remains challenging due to its high imaging cost and privacy restrictions. In our pilot study, we advocated bringing a wealth of 2D images like chest X-rays as compensation for the lack of 3D data, aiming to build a universal medical self-supervised representation learning framework, called UniMiSS. Especially, we designed a pyramid U- like medical Transformer (MiT) as the backbone to make UniMiSS possible to perform SSL with both 2D and 3D images. Consequently, the predecessor UniMiSS has two obvious merits compared to current 3D-specific SSL: (1) more effective - superior to learning strong representations, benefiting from more and diverse data; and (2) more versatile - suitable for various downstream tasks without the restriction on the dimensionality barrier. Unfortunately, UniMiSS did not dig deeply into the intrinsic anatomy correlation between 2D medical images and 3D volumes due to the lack of paired multi-modal/dimension patient data. In this extension paper, we propose the UniMiSS+, in which we introduce the digitally reconstructed radiographs (DRR) technology to simulate X-ray images from a CT volume to access paired CT and X-ray data. Benefiting from the paired group, we introduce an extra pair- wise constraint to boost the cross-modality correlation learning, which also can be adopted as a cross-dimension regularization to further improve the representations. We conduct expensive experiments on multiple 3D/2D medical image analysis tasks, including segmentation and classification. The results show that the proposed UniMiSS+ achieves promising performance on various downstream tasks, not only outperforming the ImageNet pre-training and other advanced SSL counterparts substantially but also improving the predecessor UniMiSS pre-training. Code is available at: https://github.com/YtongXie/UniMiSS-code.

CT image-based 3D inflorescence estimation of Chrysanthemum seticuspe.

To study plant organs, it is necessary to investigate the three-dimensional (3D) structures of plants. In recent years, non-destructive measurements through computed tomography (CT) have been used to understand the 3D structures of plants. In this study, we use the Chrysanthemum seticuspe capitulum inflorescence as an example and focus on contact points between the receptacles and florets within the 3D capitulum inflorescence bud structure to investigate the 3D arrangement of the florets on the receptacle. To determine the 3D order of the contact points, we constructed slice images from the CT volume data and detected the receptacles and florets in the image. However, because each CT sample comprises hundreds of slice images to be processed and each C. seticuspe capitulum inflorescence comprises several florets, manually detecting the receptacles and florets is labor-intensive. Therefore, we propose an automatic contact point detection method based on CT slice images using image recognition techniques. The proposed method improves the accuracy of contact point detection using prior knowledge that contact points exist only around the receptacle. In addition, the integration of the detection results enables the estimation of the 3D position of the contact points. According to the experimental results, we confirmed that the proposed method can detect contacts on slice images with high accuracy and estimate their 3D positions through clustering. Additionally, the sample-independent experiments showed that the proposed method achieved the same detection accuracy as sample-dependent experiments.

Research Report on the Consistency of Radiation Dose Structured Report (RDSR)

In 2020, recording and management of exposure doses for computed tomography (CT) became mandatory. In terms of dose management, information regarding the imaged body part is particularly important information for proper tabulation. However, the actual body part to be imaged and the body part information obtained from the device may differ. In this study, we investigated the difference between the imaged area information obtained from a CT device and the dose index when the area is divided into actual areas to be examined using a facility's unique imaging protocol. We collected 734784 radiation dose structured reports (RDSR) examined from 2014 to 2021 on CT equipment at 8 facilities and analyzed each facility's volume CT dose index (CTDIvol) and dose length product (DLP). The median values were tabulated and compared. Pre- and post-classing the body part, for CTDIvol, increasing tendency was observed only in the head and abdomen. A similar trend was observed for DLP. Regarding dose management using RDSR as an information source, there were no major differences in dose information depending on the classification of the body part to be imaged. We hope that the accuracy of dose management will be improved by accurately classifying the body parts to be imaged.

Patient-specific Effective Dose Estimation from Dose-Length Product in Lung Computed Tomography Using Monte Carlo Simulation.

Computed tomography (CT) imaging has a large portion in the dose of patients from radiological procedures; therefore, accurate calculation of radiation risk estimation in this modality is inevitable. In this study, a method for determining the patient-specific effective dose using the dose-length product (DLP) index in lung CT scan using Monte Carlo (MC) simulation is introduced. EGSnrc/BEAMnrc MC code was used to simulate a CT scanner. The DOSxyznrc simulation code was used to simulate a specific voxelized phantom from the patient's lungs and irradiate it according to X-ray parameter of routing lung CT scan, and dose delivered to thorax organs was calculated. Three types of phantoms were simulated according to three different body habits (slim, standard, and fat patients) in two groups of men and women. A factor was used to convert the relative dose per particle in MC code to the absolute dose. The dose was calculated in all lung organs, and the effective dose was calculated for all three groups of patient body habits. DLP index and volume CT dose index (CTDIvol) were extracted from the patient's dose report in the CT scanner. The DLP to effective dose conversion factor (k-factor) for patients with different body habitus was calculated. Lung radiation dose in slim, standard, and fat patients in men was 0.164, 0.103, and 0.078 mGy/mAs and in women was 0.164, 0.105, and 0.079 mGy/mAs, respectively. The k-factor in the group of slim patients, especially in women, was higher than in other groups. CT scan dose indexes for slim patients are reported to be underestimated in studies. The dose report in CT scan systems should be modified in proportion to the patient's body habitus, to accurately estimate the radiation risk.

Evaluation of radiation dose and establishment of local DRLs for adult during abdominopelvic CT scan imaging for university hospital centers, Morocco

Objective: In this study, we have planned to establish the local Diagnostic Reference Levels (DRLs) for the five University Hospital Centers in Morocco and to propose a national DRL for abdominopelvic CT examinations for adult patients. Materials and methods: Data were collected using a specially designed form, which included patient demographics, exposure parameters and dosimetric indicators such as dose-length product (DLP) and CT volume dose index (CTDIvol). The establishment of local and national DRL in terms of CTDIvol and DLP was carried out in accordance with the methodology described in Report 135 of the International Commission on Radiological Protection (ICRP). The effective dose was assessed using the DLP and the dose conversion factor appropriate for this location. Results: A total of 300 patients undergoing abdominopelvic examinations at the five Moroccan university hospitals were collected for this study. Patients' mean age, weight and BMI were (50.97 ± 15.05) years (67.94 ± 11.41) Kg and (25.71 ± 3.03) Kg/m2 respectively. The local DRL, in terms of DLP were 551.56, 2293.33, 709.02, 843.52 and 1059.62 mGy-cm and in terms of CTDIvol were 11,05, 46,40, 13,14, 12,01, 9,88 mGy respectively for UHC-1, UHC-2, UHC-3, UHC-4 and UHC-5 hospitals. The proposed National DRL, defined as a single value for the 75th percentile on the aggregation of data from the five centers, for abdominopelvic scan CT was 773.20 mGy.cm and 12.05 mGy for DLP and CTDIvol, respectively. Conclusion: The Moroccan DRL was comparable to those of other countries, ranking 10th among 21 countries. However, we are still seeking to reduce and optimize radiation dose while maintaining good image quality for reliable diagnosis, and this can be achieved by enhancing radiographers' and radiologists' knowledge of radiation protection, protocol justification, and optimization through rigorous hospital radiation audits at national level.

McTFI QSM MRI ABC/2 intracranial hemorrhage to noncontrast head CT volume measurement equivalence

Background/objectivesIntracranial hemorrhage (ICH) volume assessment is an important part of patient management and is routinely obtained by non-contrast head CT (NCHCT) using the validated ABC/2 measurement method. Because conventional MRI imaging sequences demonstrate variability in ICH appearance, volumetric analyses for MRI bleed volume in a standardized manner using ABC/2 is not possible. The recently introduced multiecho-complex total field inversion quantitative susceptibility mapping (mcTFI QSM) MRI technique, which maps brain tissue susceptibility to both depict brain tissue structures and quantify tissue susceptibility, may provide a viable alternative. In this study we evaluated mcTFI QSM ABC/2 ICH volume assessment relative to NCHCT. MethodsPatients with ICH who had undergone NCHCT and MRI brain scans within 48 h were recruited for this retrospective study. The ABC/2 method was applied to estimate the bleed volume for both NCHCT and MRI by a CAQ-certified neuroradiologist with 10 years of experience and a trained laboratory assistant. Results were analyzed via Bland-Altman (B-A) and linear regression. Results54 patients (27 females) who had undergone NCHCT and MRI within 48 h (<24 h., n = 31, 24-48 h, n = 10) were enrolled. mcTFI QSM ICH volume measurement method showed a positive correlation (99.5%) compared to NCHCT. B-A plot comparing ABC/2 ICH volume on NCHCT and mcTFI MRI done for patients within 24 h demonstrates a bias of −0.09%. ConclusionsICH volume calculation using ABC/2 on mcTFI QSM showed a high correlation with NCHCT measurement. These results suggest mcTFI QSM is a promising MRI method for ABC/2 for bleed volume measurement.

Improving Sonication Efficiency in Transcranial MR-Guided Focused Ultrasound Treatment: A Patient-Data Simulation Study.

The potential improvement in sonication efficiency achieved by tilting the focused ultrasound (FUS) transducer of the transcranial MR-guided FUS system is presented. A total of 56 cases of patient treatment data were used. The relative position of the clinical FUS transducer to the patient's head was reconstructed, and region-specific skull density and porosity were calculated based on the patient's CT volume image. The total transmission coefficient of acoustic waves emitted from each channel was calculated. Then, the total energy penetrating the human skull-which represents the sonication efficiency-was estimated. As a result, improved sonication efficiency was by titling the FUS transducer to a more appropriate angle achieved in all 56 treatment cases. This simulation result suggests the potential improvement in transcranial-focused ultrasound treatment by simply adjusting the transducer angle.

Multi-Modal Ensemble Deep Learning in Head and Neck Cancer HPV Sub-Typing.

Oropharyngeal Squamous Cell Carcinoma (OPSCC) is one of the common forms of heterogeneity in head and neck cancer. Infection with human papillomavirus (HPV) has been identified as a major risk factor for OPSCC. Therefore, differentiating the HPV-positive and negative cases in OPSCC patients is an essential diagnostic factor influencing future treatment decisions. In this study, we investigated the accuracy of a deep learning-based method for image interpretation and automatically detected the HPV status of OPSCC in routinely acquired Computed Tomography (CT) and Positron Emission Tomography (PET) images. We introduce a 3D CNN-based multi-modal feature fusion architecture for HPV status prediction in primary tumor lesions. The architecture is composed of an ensemble of CNN networks and merges image features in a softmax classification layer. The pipeline separately learns the intensity, contrast variation, shape, texture heterogeneity, and metabolic assessment from CT and PET tumor volume regions and fuses those multi-modal features for final HPV status classification. The precision, recall, and AUC scores of the proposed method are computed, and the results are compared with other existing models. The experimental results demonstrate that the multi-modal ensemble model with soft voting outperformed single-modality PET/CT, with an AUC of 0.76 and F1 score of 0.746 on publicly available TCGA and MAASTRO datasets. In the MAASTRO dataset, our model achieved an AUC score of 0.74 over primary tumor volumes of interest (VOIs). In the future, more extensive cohort validation may suffice for better diagnostic accuracy and provide preliminary assessment before the biopsy.

Benefits of low-dose carotid CT angiography in stroke patients

Introduction: Computed tomography angiography (CTA) represents the gold standard as a method for the diagnosis of carotid artery diseases. The current topic is the use of CTA for the evaluation of carotid arteries with a reduction in the dose of contrast agent and dose of ionizing radiation, which, with adequate preparation, would enable the use of this method in some risk groups. The aim of this study was to evaluate the feasibility and image quality of a new low-dose CTA protocol in comparison with a standard protocol. Methods: Forty patients with recumbent ischemic stroke were included in the study, twenty of whom underwent lowdose CTA, and the remaining twenty underwent a standard CTA protocol of the carotid arteries. Results: No significant difference was found between the mean values of CT number (Hounsfield unit), signal-to-noise ratio, contrast-to-noise ratio, and subjective assessment of image quality in the comparison of the control and experimental groups. CT dose index, volume, and dose length product were significantly lower in patients who underwent lowdose carotid CTA. There was no significant difference in the degree of carotid stenosis between color Doppler and CTA. Conclusion: The use of the low-dose protocol for carotid CTA allows the application of this method in risk groups, in which it was previously not possible to perform, with the same image quality in comparison with the standard protocol.

Predicting subsolid pulmonary nodules before percutaneous needle biopsy: a comparison of artificial neural network and biopsy results

To establish an artificial neural network (ANN) model to predict subsolid nodules (SSNs) before percutaneous core-needle biopsy (PCNB). The results of the two methods were compared to provide guidance on the treatment of SSNs. This was a single-centre retrospective study using data from 1,459 SSNs between 2013 and 2021. The ANN was developed using data from patients who underwent surgery following computed tomography (CT) (SFC) and validated using data from patients who underwent surgery following biopsy (SFB). The prediction results of the ANN for the PCNB group and the histopathological results obtained after biopsy were compared with the histopathological results of lung nodules in the same group after surgery. Additionally, the choice of predictors for PCNB was analysed using multivariate analysis. There was no significant difference between the accuracies of the ANN and PCNB in the SFB group (p=0.086). The sensitivity of PCNB was lower than that of the ANN (p=0.000), but the specificity was higher (p=0.001). PCNB had better diagnostic ability than the ANN. The incidence of precursor lesions and non-neoplastic lesions in the SFB group was lower than that in the SFC group (p=0.000). A history of malignant tumours, size (2-3 cm), volume (>400 cm3) and mean CT value (≥-450 HU) are important factors for selecting PCNB. Both ANN and PCNB have comparable accuracy in diagnosing SSNs; however, PCNB has a slightly higher diagnostic ability than ANN. Selecting appropriate patients for PCNB is important for maximising the benefit to SSN patients.

3D printing for an anterolateral thigh phalloplasty

BackgroundPhalloplasty procedures are performed to create a phallus, typically as a gender-affirming surgery for treating gender dysphoria. Due to the controversial nature of this specific procedure, more innovation is needed to directly assist surgical teams in this field. As a result, surgeons are left to improvise and adapt tools created for other procedures to improve surgical outcomes. This study developed a patient-specific 3D printed model from segmented computed tomography (CT) scans to accurately represent the relevant vasculature necessary for anterolateral thigh (ALT) flap phalloplasty. The surgical procedure seeks to maintain intact vessels that derive from the descending branch of the lateral circumflex femoral artery, typically found traveling within the intermuscular septum between the rectus femoris and vastus lateralis.MethodsIn this study, we created and printed 3D models of the leg and vasculature using two techniques: (1) a standard segmentation technique with the addition of a reference grid and (2) a bitmap method in which the total CT volume is colorized and printed.ResultsThe results gathered included the physician’s view on the model’s accuracy and visualization of relevant anatomy. Bitmap-printed models resulted in a high amount of detail, eliciting surgeons’ undesirable reactions due to the excess of information. The hybrid method produced favorable results, indicating positive feasibility.ConclusionsThis study tested the ability to accurately print a patient-specific 3D model that could represent the vasculature necessary for ALT flap procedures and potentially be used in surgical reference and planning in the future. A surgeon performing phalloplasty procedures discussed their approval of both models and their preference for grid creation and application.

An evaluation of CT radiation doses within the Yorkshire Lung Screening Trial.

To evaluate radiation doses for all low-dose CT scans performed during the first year of a lung screening trial. For all lung screening scans that were performed using a CT protocol that delivered image quality meeting the RSNA QIBA criteria, radiation dose metrics, participant height, weight, gender, and age were recorded. Values of volume CT dose index (CTDIvol) and dose length product (DLP) were evaluated as a function of weight in order to assess the performance of the scan protocol across the participant cohort. Calculated effective doses were used to establish the additional lifetime attributable cancer risks arising from trial scans. Median values of CTDIvol, DLP, and effective dose (IQR) from the 3521 scans were 1.1mGy (0.70), 42.4mGycm (24.9), and 1.15mSv (0.67), whilst for 60-80kg participants the values were 1.0mGy (0.30), 35.8mGycm (11.4), and 0.97mSv (0.31). A statistically significant correlation between CTDIvol and weight was identified for males (r = 0.9123, P < .001) and females (r = 0.9052, P < .001), however, the effect of gender on CTDIvol was not statistically significant (P = .2328) despite notable differences existing at the extremes of the weight range. The additional lifetime attributable cancer risks from a single scan were in the range 0.001%-0.006%. Low radiation doses can be achieved across a typical lung screening cohort using scan protocols that have been shown to deliver high levels of image quality. The observed dose levels may be considered as typical values for lung screening scans on similar types of scanners for an equivalent participant cohort. Presentation of typical radiation dose levels for CT lung screening examinations in a large UK trial. Effective radiation doses can be of the order of 1mSv for standard sized participants. Lifetime attributable cancer risks resulting from a single low-dose CT scan did not exceed 0.006%.

Cost-effectiveness of volume computed tomography in lung cancer screening: a cohort simulation based on Nelson study outcomes

Objectives This study aimed to evaluate the cost-effectiveness of lung cancer screening (LCS) with volume-based low-dose computed tomography (CT) versus no screening for an asymptomatic high-risk population in the United Kingdom (UK), utilising the long-term insights provided by the NELSON study, the largest European randomized control trial investigating LCS. Methods A cost-effectiveness analysis was conducted using a decision tree and a state-transition Markov model to simulate the identification, diagnosis, and treatments for a lung cancer high-risk population, from a UK National Health Service (NHS) perspective. Eligible participants underwent annual volume CT screening and were compared to a cohort without the option of screening. Screen-detected lung cancers, costs, quality-adjusted life years (QALYs), and the incremental cost-effectiveness ratio (ICER) were predicted. Results Annual volume CT screening of 1.3 million eligible participants resulted in 96,474 more lung cancer cases detected in early stage, and 73,825 fewer cases in late stage, leading to 53,732 premature lung cancer deaths averted and 421,647 QALYs gained, compared to no screening. The ICER was £5,455 per QALY. These estimates were robust in sensitivity analyses. Limitations Lack of long-term survival data for lung cancer patients; deficiency in rigorous micro-costing studies to establish detailed treatment costs inputs for lung cancer patients. Conclusions Annual LCS with volume-based low-dose CT for a high-risk asymptomatic population is cost-effective in the UK, at a threshold of £20,000 per QALY, representing an efficient use of NHS resources with substantially improved outcomes for lung cancer patients, as well as additional societal and economic benefits for society as a whole. These findings advocate evidence-based decisions for the potential implementation of a nationwide LCS in the UK.

Cardiac computed tomography angiography with and without bolus tracking methods in infants with congenital heart disease.

The study investigated radiation dose, vascular computed tomography (CT) enhancement and image quality of cardiac computed tomography angiography (CCTA) with and without bolus tracking (BT) methods in infants with congenital heart disease (CHD). The volume CT dose index (CTDIvol) and dose length product (DLP) were recorded for all CT scans, and the effective dose was obtained using a conversion factors. The CT number for the ascending aorta (AO) and pulmonary artery (PA), image noise of muscle tissue and contrast-to-noise ratio (CNR) were measured and calculated. The median values in the groups with and without BT were 2.20mGy versus 0.44mGy for CTDIvol, 8.10mGy·cm versus 6.20mGy·cm for DLP, and 0.66mSv versus 0.51mSv for effective dose (p<0.001). There were no statistical differences in vascular CT enhancement, image noise, and CNR. CCTA without BT methods can reduce the radiation dose while maintaining vascular CT enhancement and image quality compared to CCTA with BT methods.

Clinical Impact of a Protocol Involving Cone-Beam CT (CBCT), Fusion Imaging and Ablation Volume Prediction in Percutaneous Image-Guided Microwave Ablation in Patients with Hepatocellular Carcinoma Unsuitable for Standard Ultrasound (US) Guidance.

to evaluate the clinical impact of a protocol for the image-guided percutaneous microwave ablation (MWA) of hepatocellular carcinoma (HCC) that includes cone-beam computed tomography (CBCT), fusion imaging and ablation volume prediction in patients with hepatocellular carcinoma unsuitable for standard ultrasound (US) guidance. this study included all patients with HCC treated with MWA between January 2021 and June 2022 in a tertiary institution. Patients were divided into two groups: Group A, treated following the protocol, and Group B, treated with standard ultrasound (US) guidance. Follow-up images were reviewed to assess residual disease (RD), local tumor progression (LTP) and intrahepatic distant recurrence (IDR). Ablation response at 1 month was also evaluated according to mRECIST. Baseline variables and outcomes were compared between the groups. For 1-month RD, propensity score weighting (PSW) was performed. 80 consecutive patients with 101 HCCs treated with MWA were divided into two groups. Group A had 41 HCCs in 37 patients, and Group B had 60 HCCs in 43 patients. Among all baseline variables, the groups differed regarding their age (mean of 72 years in Group A and 64 years in Group B, respectively), new vs. residual tumor rates (48% Group A vs. 25% Group B, p < 0.05) and number of subcapsular tumors (56.7% Group B vs. 31.7% Group A, p < 0.05) and perivascular tumors (51.7% Group B vs. 17.1% Group A, p < 0.05). The protocol led to repositioning the antenna in 49% of cases. There was a significant difference in 1-month local response between the groups measured as the RD rate and mRECIST outcomes. LTP rates at 3 and 6 months, and IDR rates at 1, 3 and 6 months, showed no significant differences. Among all variables, logistic regression after PSW demonstrated a protective effect of the protocol against 1-month RD. The use of CBCT, fusion imaging and ablation volume prediction during percutaneous MWA of HCCs provided a better 1-month tumor local control. Further studies with a larger population and longer follow-up are needed.

Radiation exposure and cancer risk of pediatric head CT scans: Impact of age and scanning parameters

Computed tomography (CT) is a widely utilized imaging technique in various medical applications due to the advanced technology that allows the diagnosis of diseases efficiently. The aims of this study were (1) to assess the radiation dose and the probability of cancer incidence for pediatric head CT scans, and (2) to analyze the influence of age and other scanning parameters on radiation dose. A total of 224 pediatric patients, ranging in age from newborn to 18 years, were collected from three main hospitals in Saudi Arabia. The patient data were divided into four age groups: Group 1 (0 to < 1 year old), Group 2 (1 to < 5 years old), Group 3 (5 to < 10 years old), and Group 4 (10 to < 18 years old). Initially, descriptive statistics were conducted for the values of volume CT dose index (CTDIvol), dose-length product (DLP), effective dose, and probability of lifetime cancer risk. This was followed by applying the Kruskal-Wallis H and the Mann-Whitney U tests to investigate the influence of age and other scanning parameters, namely tube voltage, slice thickness, and the use of automatic tube current modulation (ATCM), on radiation dose. It has been found that as the age of patients increased, there was a corresponding increase in CTDIvol and DLP values and a decrease in effective doses and lifetime cancer risk. The patient's age accounted for 33.4% and 34.9% of the variations in CTDIvol and DLP, respectively, while tube voltage (kVp) settings accounted for 25.8% and 24.9%, respectively. There was an increase of 22.3% in CTDIvol reported for patients scanned with a slice thickness of 2 mm compared to those scanned at 2.5 mm, and a 2.2% dose increase between slice thicknesses of 2.5 mm and 5 mm. Radiation doses in the surveyed hospitals fall within acceptable limits. Further studies in Saudi Arabia should focus on evaluating patient doses based on the actual sizes, which determine radiation doses for pediatric CT imaging more accurately.

Immune Checkpoint Inhibitor-Induced Pancreatic Injury: Clinical and Radiological Profile and Response to Steroids

Background and AimsImmune checkpoint inhibitor therapy causes numerous immune related adverse events, including autoimmune pancreatic injury (AIPI), which results in rapid organ atrophy. We profiled the clinico-radiological features, short-term natural history, and response to steroids of AIPI. MethodsWe retrospectively reviewed medical records of 229/11,165 (2.1%) adult patients with AIPI. 110/229 (48%) had abdominal computerized tomography scan (CT) at lipase elevation; data of 110 without pancreatic metastases were analyzed. We analyzed serial CT-based pancreas volumetry data in 48 patients with AIPI (32 with normal CT and 16 with pancreatitis on CT at lipase elevation). We examined impact of steroids on pain and disease course. ResultsIn AIPI (n=229), median lipase elevation was 4x upper limit of normal (range: 3-40x). The injury was more often asymptomatic than painful (143/229 (62%) vs 86/229 (38%), P<.000). Majority (83/110 (75%) had normal CT, often in painless vs painful disease: 51/57 (90%) vs 32/53 (60%), P<.001); 25% had interstitial pancreatitis. On serial pancreas volumetry, marked volume (cc) loss occurred 1 year after vs 3 months before lipase elevation in both normal CT (median 81.6 vs 61.3, P=.00) and pancreatitis on CT groups (91.8 vs 60.5, P=.00), >20% volume loss occurred in 47% vs 73%, respectively (P=.08). Steroids, when used did not mitigate pain, biochemical relapse, pancreas volume loss or 1-year diabetes incidence (7.2%). ConclusionsAIPI is uniquely characterized by painless lipase elevation, normal pancreas on CT and rapid pancreatic volume loss on follow-up. Steroids do not appear to have a role in management.

Relationship between Thyroid CT Density, Volume, and Future TSH Elevation: A 5-Year Follow-Up Study.

This study aimed to explore the relationship between thyroid-stimulating hormone (TSH) elevation and the baseline computed tomography (CT) density and volume of the thyroid. We examined 86 cases with new-onset hypothyroidism (TSH > 4.5 IU/mL) and 1071 controls from a medical check-up database over 5 years. A deep learning-based thyroid segmentation method was used to assess CT density and volume. Statistical tests and logistic regression were employed to determine differences and odds ratios. Initially, the case group showed a higher CT density (89.8 vs. 81.7 Hounsfield units (HUs)) and smaller volume (13.0 vs. 15.3 mL) than those in the control group. For every +10 HU in CT density and -3 mL in volume, the odds of developing hypothyroidism increased by 1.40 and 1.35, respectively. Over the course of the study, the case group showed a notable CT density reduction (median: -8.9 HU), whereas the control group had a minor decrease (-2.9 HU). Thyroid volume remained relatively stable for both groups. Higher CT density and smaller thyroid volume at baseline are correlated with future TSH elevation. Over time, there was a substantial and minor decrease in CT density in the case and control groups, respectively. Thyroid volumes remained consistent in both cohorts.

PST-Radiomics: a PET/CT lymphoma classification method based on pseudo spatial-temporal radiomic features and structured atrous recurrent convolutional neural network

Objective. Existing radiomic methods tend to treat each isolated tumor as an inseparable whole, when extracting radiomic features. However, they may discard the critical intra-tumor metabolic heterogeneity (ITMH) information, that contributes to triggering tumor subtypes. To improve lymphoma classification performance, we propose a pseudo spatial-temporal radiomic method (PST-Radiomics) based on positron emission tomography computed tomography (PET/CT). Approach. Specifically, to enable exploitation of ITMH, we first present a multi-threshold gross tumor volume sequence (GTVS). Next, we extract 1D radiomic features based on PET images and each volume in GTVS and create a pseudo spatial-temporal feature sequence (PSTFS) tightly interwoven with ITMH. Then, we reshape PSTFS to create 2D pseudo spatial-temporal feature maps (PSTFM), of which the columns are elements of PSTFS. Finally, to learn from PSTFM in an end-to-end manner, we build a light-weighted pseudo spatial-temporal radiomic network (PSTR-Net), in which a structured atrous recurrent convolutional neural network serves as a PET branch to better exploit the strong local dependencies in PSTFM, and a residual convolutional neural network is used as a CT branch to exploit conventional radiomic features extracted from CT volumes. Main results. We validate PST-Radiomics based on a PET/CT lymphoma subtype classification task. Experimental results quantitatively demonstrate the superiority of PST-Radiomics, when compared to existing radiomic methods. Significance. Feature map visualization of our method shows that it performs complex feature selection while extracting hierarchical feature maps, which qualitatively demonstrates its superiority.

ORRN: An ODE-Based Recursive Registration Network for Deformable Respiratory Motion Estimation With Lung 4DCT Images.

Deformable Image Registration (DIR) plays a significant role in quantifying deformation in medical data. Recent Deep Learning methods have shown promising accuracy and speedup for registering a pair of medical images. However, in 4D (3D + time) medical data, organ motion, such as respiratory motion and heart beating, can not be effectively modeled by pair-wise methods as they were optimized for image pairs but did not consider the organ motion patterns necessary when considering 4D data. This article presents ORRN, an Ordinary Differential Equations (ODE)-based recursive image registration network. Our network learns to estimate time-varying voxel velocities for an ODE that models deformation in 4D image data. It adopts a recursive registration strategy to progressively estimate a deformation field through ODE integration of voxel velocities. We evaluate the proposed method on two publicly available lung 4DCT datasets, DIRLab and CREATIS, for two tasks: 1) registering all images to the extreme inhale image for 3D+t deformation tracking and 2) registering extreme exhale to inhale phase images. Our method outperforms other learning-based methods in both tasks, producing the smallest Target Registration Error of 1.24 mm and 1.26 mm, respectively. Additionally, it produces less than 0.001% unrealistic image folding, and the computation speed is less than 1 s for each CT volume. ORRN demonstrates promising registration accuracy, deformation plausibility, and computation efficiency on group-wise and pair-wise registration tasks. It has significant implications in enabling fast and accurate respiratory motion estimation for treatment planning in radiation therapy or robot motion planning in thoracic needle insertion.