Lung vessel connectivity map as anatomical prior knowledge for deep learning-based lung lobe segmentation.

Purpose: Our study investigates the potential benefits of incorporating prior anatomical knowledge into a deep learning (DL) method designed for the automated segmentation of lung lobes in chest CT scans. Approach: We introduce an automated DL-based approach that leverages anatomical information from the lung's vascular system to guide and enhance the segmentation process. This involves utilizing a lung vessel connectivity (LVC) map, which encodes relevant lung vessel anatomical data. Our study explores the performance of three different neural network architectures within the nnU-Net framework: a standalone U-Net, a multitasking U-Net, and a cascade U-Net. Results: Experimental findings suggest that the inclusion of LVC information in the DL model can lead to improved segmentation accuracy, particularly, in the challenging boundary regions of expiration chest CT volumes. Furthermore, our study demonstrates the potential for LVC to enhance the model's generalization capabilities. Finally, the method's robustness is evaluated through the segmentation of lung lobes in 10 cases of COVID-19, demonstrating its applicability in the presence of pulmonary diseases. Conclusions: Incorporating prior anatomical information, such as LVC, into the DL model shows promise for enhancing segmentation performance, particularly in the boundary regions. However, the extent of this improvement has limitations, prompting further exploration of its practical applicability.

Automated region growing-based segmentation for trabecular bone structure in fresh-frozen human wrist specimens.

Bone strength depends on both mineral content and bone structure. Measurements of bone microstructure on specimens can be performed by micro-CT. In vivo measurements are reliably performed by high-resolution peripheral computed tomography (HR-pQCT) using dedicated software. In previous studies from our research group, trabecular bone properties on CT data of defatted specimens from many different CT devices have been analyzed using an Automated Region Growing (ARG) algorithm-based code, showing strong correlations to micro-CT.The aim of the study was to validate the possibility of segmenting and measuring trabecular bone structure from clinical CT data of fresh-frozen human wrist specimens. Data from micro-CT was used as reference. The hypothesis was that the ARG-based in-house built software could be used for such measurements.HR-pQCT image data at two resolutions (61 and 82 µm isotropic voxels) from 23 fresh-frozen human forearms were analyzed. Correlations to micro-CT were strong, varying from 0.72 to 0.99 for all parameters except trabecular termini and nodes. The bone volume fraction had correlations varying from 0.95 to 0.98 but was overestimated compared to micro-CT, especially at the lower resolution. Trabecular separation and spacing were the most stable parameters with correlations at 0.80-0.97 and mean values in the same range as micro-CT.Results from this in vitro study show that an ARG-based software could be used for segmenting and measuring 3D trabecular bone structure from clinical CT data of fresh-frozen human wrist specimens using micro-CT data as reference. Over-and underestimation of several of the bone structure parameters must however be taken into account.

Pulmonary volumes and signs of chronic airflow limitation in quantitative computed tomography.

BACKGROUND: Computed tomography (CT) offers pulmonary volumetric quantification but is not commonly used in healthy individuals due to radiation concerns. Chronic airflow limitation (CAL) is one of the diagnostic criteria for chronic obstructive pulmonary disease (COPD), where early diagnosis is important. Our aim was to present reference values for chest CT volumetric and radiodensity measurements and explore their potential in detecting early signs of CAL. METHODS: From the population-based Swedish CArdioPulmonarybioImage Study (SCAPIS), 294 participants aged 50-64, were categorized into non-CAL (n = 258) and CAL (n = 36) groups based on spirometry. From inspiratory and expiratory CT images we compared lung volumes, mean lung density (MLD), percentage of low attenuation volume (LAV%) and LAV cluster volume between groups, and against reference values from static pulmonary function test (PFT). RESULTS: The CAL group exhibited larger lung volumes, higher LAV%, increased LAV cluster volume and lower MLD compared to the non-CAL group. Lung volumes significantly deviated from PFT values. Expiratory measurements yielded more reliable results for identifying CAL compared to inspiratory. Using a cut-off value of 0.6 for expiratory LAV%, we achieved sensitivity, specificity and positive/negative predictive values of 72%, 85% and 40%/96%, respectively. CONCLUSION: We present volumetric reference values from inspiratory and expiratory chest CT images for a middle-aged healthy cohort. These results are not directly comparable to those from PFTs. Measures of MLD and LAV can be valuable in the evaluation of suspected CAL. Further validation and refinement are necessary to demonstrate its potential as a decision support tool for early detection of COPD.

Acquisition Duration Optimization Using Visual Grading Regression in [68Ga]FAPI-46 PET Imaging of Oncologic Patients.

Fibroblast activation protein is a promising target for oncologic molecular imaging with radiolabeled fibroblast activation protein inhibitors (FAPI) in a large variety of cancers. However, there are yet no published recommendations on how to set up an optimal imaging protocol for FAPI PET/CT. It is important to optimize the acquisition duration and strive toward an acquisition that is sufficiently short while simultaneously providing sufficient image quality to ensure a reliable diagnosis. The aim of this study was to evaluate the feasibility of reducing the acquisition duration of [68Ga]FAPI-46 imaging while maintaining satisfactory image quality, with certainty that the radiologist's ability to make a clinical diagnosis would not be affected. Methods: [68Ga]FAPI-46 PET/CT imaging was performed on 10 patients scheduled for surgical resection of suspected pancreatic cancer, 60 min after administration of 3.6 ± 0.2 MBq/kg. The acquisition time was 4 min/bed position, and the raw PET data were statistically truncated and reconstructed to represent images with an acquisition duration of 1, 2, and 3 min/bed position, additional to the reference images of 4 min/bed position. Four image quality criteria that focused on the ability to distinguish specific anatomic details, as well as perceived image noise and overall image quality, were scored on a 4-point Likert scale and analyzed with mixed-effects ordinal logistic regression. Results: A trend toward increasing image quality scores with increasing acquisition duration was observed for all criteria. For the overall image quality, there was no significant difference between 3 and 4 min/bed position, whereas 1 and 2 min/bed position were rated significantly (P < 0.05) lower than 4 min/bed position. For the other criteria, all images with a reduced acquisition duration were rated significantly inferior to images obtained at 4 min/bed position. Conclusion: The acquisition duration can be reduced from 4 to 3 min/bed position while maintaining satisfactory image quality. Reducing the acquisition duration to 2 min/bed position or lower is not recommended since it results in inferior-quality images so noisy that clinical interpretation is significantly disrupted.

Image quality in CT thorax: effect of altering reconstruction algorithm and tube load.

Non-linear properties of iterative reconstruction (IR) algorithms can alter image texture. We evaluated the effect of a model-based IR algorithm (advanced modelled iterative reconstruction; ADMIRE) and dose on computed tomography thorax image quality. Dual-source scanner data were acquired at 20, 45 and 65 reference mAs in 20 patients. Images reconstructed with filtered back projection (FBP) and ADMIRE Strengths 3-5 were assessed independently by six radiologists and analysed using an ordinal logistic regression model. For all image criteria studied, the effects of tube load 20 mAs and all ADMIRE strengths were significant (p < 0.001) when compared to reference categories 65 mAs and FBP. Increase in tube load from 45 to 65 mAs showed image quality improvement in three of six criteria. Replacing FBP with ADMIRE significantly improves perceived image quality for all criteria studied, potentially permitting a dose reduction of almost 70% without loss in image quality.

Learning effects in visual grading assessment of model-based reconstruction algorithms in abdominal Computed Tomography.

Objectives: Images reconstructed with higher strengths of iterative reconstruction algorithms may impair radiologists' subjective perception and diagnostic performance due to changes in the amplitude of different spatial frequencies of noise. The aim of the present study was to ascertain if radiologists can learn to adapt to the unusual appearance of images produced by higher strengths of Advanced modeled iterative reconstruction algorithm (ADMIRE). Methods: Two previously published studies evaluated the performance of ADMIRE in non-contrast and contrast-enhanced abdominal CT. Images from 25 (first material) and 50 (second material) patients, were reconstructed with ADMIRE strengths 3, 5 (AD3, AD5) and filtered back projection (FBP). Radiologists assessed the images using image criteria from the European guidelines for quality criteria in CT. To ascertain if there was a learning effect, new analyses of data from the two studies was performed by introducing a time variable in the mixed-effects ordinal logistic regression model. Results: In both materials, a significant negative attitude to ADMIRE 5 at the beginning of the viewing was strengthened during the progress of the reviews for both liver parenchyma (first material: -0.70, p < 0.01, second material: -0.96, p < 0.001) and overall image quality (first material:-0.59, p < 0.05, second material::-1.26, p < 0.001). For ADMIRE 3, an early positive attitude for the algorithm was noted, with no significant change over time for all criteria except one (overall image quality), where a significant negative trend over time (-1.08, p < 0.001) was seen in the second material. Conclusions: With progression of reviews in both materials, an increasing dislike for ADMIRE 5 images was apparent for two image criteria. In this time perspective (weeks or months), no learning effect towards accepting the algorithm could be demonstrated.

Validation of automated post-adjustments of HDR prostate brachytherapy treatment plans by quantitative measures and oncologist observer study.

PURPOSE: The aim was to evaluate a postprocessing optimization algorithm's ability to improve the spatial properties of a clinical treatment plan while preserving the target coverage and the dose to the organs at risk. The goal was to obtain a more homogenous treatment plan, minimizing the need for manual adjustments after inverse treatment planning. MATERIALS AND METHODS: The study included 25 previously treated prostate cancer patients. The treatment plans were evaluated on dose-volume histogram parameters established clinical and quantitative measures of the high dose volumes. The volumes of the four largest hot spots were compared and complemented with a human observer study with visual grading by eight oncologists. Statistical analysis was done using ordinal logistic regression. Weighted kappa and Fleiss' kappa were used to evaluate intra- and interobserver reliability. RESULTS: The quantitative analysis showed that there was no change in planning target volume (PTV) coverage and dose to the rectum. There were significant improvements for the adjusted treatment plan in: V150% and V200% for PTV, dose to urethra, conformal index, and dose nonhomogeneity ratio. The three largest hot spots for the adjusted treatment plan were significantly smaller compared to the clinical treatment plan. The observers preferred the adjusted treatment plan in 132 cases and the clinical in 83 cases. The observers preferred the adjusted treatment plan on homogeneity and organs at risk but preferred the clinical plan on PTV coverage. CONCLUSIONS: Quantitative analysis showed that the postadjustment optimization tool could improve the spatial properties of the treatment plans while maintaining the target coverage.

Multi-cohort and longitudinal Bayesian clustering study of stage and subtype in Alzheimer's disease.

Understanding Alzheimer's disease (AD) heterogeneity is important for understanding the underlying pathophysiological mechanisms of AD. However, AD atrophy subtypes may reflect different disease stages or biologically distinct subtypes. Here we use longitudinal magnetic resonance imaging data (891 participants with AD dementia, 305 healthy control participants) from four international cohorts, and longitudinal clustering to estimate differential atrophy trajectories from the age of clinical disease onset. Our findings (in amyloid-ß positive AD patients) show five distinct longitudinal patterns of atrophy with different demographical and cognitive characteristics. Some previously reported atrophy subtypes may reflect disease stages rather than distinct subtypes. The heterogeneity in atrophy rates and cognitive decline within the five longitudinal atrophy patterns, potentially expresses a complex combination of protective/risk factors and concomitant non-AD pathologies. By alternating between the cross-sectional and longitudinal understanding of AD subtypes these analyses may allow better understanding of disease heterogeneity.

Photon-counting detector CT and energy-integrating detector CT for trabecular bone microstructure analysis of cubic specimens from human radius.

BACKGROUND: As bone microstructure is known to impact bone strength, the aim of this in vitro study was to evaluate if the emerging photon-counting detector computed tomography (PCD-CT) technique may be used for measurements of trabecular bone structures like thickness, separation, nodes, spacing and bone volume fraction. METHODS: Fourteen cubic sections of human radius were scanned with two multislice CT devices, one PCD-CT and one energy-integrating detector CT (EID-CT), using micro-CT as a reference standard. The protocols for PCD-CT and EID-CT were those recommended for inner- and middle-ear structures, although at higher mAs values: PCD-CT at 450 mAs and EID-CT at 600 (dose equivalent to PCD-CT) and 1000 mAs. Average measurements of the five bone parameters as well as dispersion measurements of thickness, separation and spacing were calculated using a three-dimensional automated region growing (ARG) algorithm. Spearman correlations with micro-CT were computed. RESULTS: Correlations with micro-CT, for PCD-CT and EID-CT, ranged from 0.64 to 0.98 for all parameters except for dispersion of thickness, which did not show a significant correlation (p = 0.078 to 0.892). PCD-CT had seven of the eight parameters with correlations ρ > 0.7 and three ρ > 0.9. The dose-equivalent EID-CT instead had four parameters with correlations ρ > 0.7 and only one ρ > 0.9. CONCLUSIONS: In this in vitro study of radius specimens, strong correlations were found between trabecular bone structure parameters computed from PCD-CT data when compared to micro-CT. This suggests that PCD-CT might be useful for analysing bone microstructure in the peripheral human skeleton.

A review on AI-based medical image computing in head and neck surgery.

Head and neck surgery is a fine surgical procedure with a complex anatomical space, difficult operation and high risk. Medical image computing (MIC) that enables accurate and reliable preoperative planning is often needed to reduce the operational difficulty of surgery and to improve patient survival. At present, artificial intelligence, especially deep learning, has become an intense focus of research in MIC. In this study, the application of deep learning-based MIC in head and neck surgery is reviewed. Relevant literature was retrieved on the Web of Science database from January 2015 to May 2022, and some papers were selected for review from mainstream journals and conferences, such as IEEE Transactions on Medical Imaging, Medical Image Analysis, Physics in Medicine and Biology, Medical Physics, MICCAI, etc. Among them, 65 references are on automatic segmentation, 15 references on automatic landmark detection, and eight references on automatic registration. In the elaboration of the review, first, an overview of deep learning in MIC is presented. Then, the application of deep learning methods is systematically summarized according to the clinical needs, and generalized into segmentation, landmark detection and registration of head and neck medical images. In segmentation, it is mainly focused on the automatic segmentation of high-risk organs, head and neck tumors, skull structure and teeth, including the analysis of their advantages, differences and shortcomings. In landmark detection, the focus is mainly on the introduction of landmark detection in cephalometric and craniomaxillofacial images, and the analysis of their advantages and disadvantages. In registration, deep learning networks for multimodal image registration of the head and neck are presented. Finally, their shortcomings and future development directions are systematically discussed. The study aims to serve as a reference and guidance for researchers, engineers or doctors engaged in medical image analysis of head and neck surgery.

Prior-aware autoencoders for lung pathology segmentation.

Segmentation of lung pathology in Computed Tomography (CT) images is of great importance for lung disease screening. However, the presence of different types of lung pathologies with a wide range of heterogeneities in size, shape, location, and texture, on one side, and their visual similarity with respect to surrounding tissues, on the other side, make it challenging to perform reliable automatic lesion segmentation. To leverage segmentation performance, we propose a deep learning framework comprising a Normal Appearance Autoencoder (NAA) model to learn the distribution of healthy lung regions and reconstruct pathology-free images from the corresponding pathological inputs by replacing the pathological regions with the characteristics of healthy tissues. Detected regions that represent prior information regarding the shape and location of pathologies are then integrated into a segmentation network to guide the attention of the model into more meaningful delineations. The proposed pipeline was tested on three types of lung pathologies, including pulmonary nodules, Non-Small Cell Lung Cancer (NSCLC), and Covid-19 lesion on five comprehensive datasets. The results show the superiority of the proposed prior model, which outperformed the baseline segmentation models in all the cases with significant margins. On average, adding the prior model improved the Dice coefficient for the segmentation of lung nodules by 0.038, NSCLCs by 0.101, and Covid-19 lesions by 0.041. We conclude that the proposed NAA model produces reliable prior knowledge regarding the lung pathologies, and integrating such knowledge into a prior segmentation network leads to more accurate delineations.

Spherical Convolutional Neural Networks for Survival Rate Prediction in Cancer Patients.

Objective: Survival Rate Prediction (SRP) is a valuable tool to assist in the clinical diagnosis and treatment planning of lung cancer patients. In recent years, deep learning (DL) based methods have shown great potential in medical image processing in general and SRP in particular. This study proposes a fully-automated method for SRP from computed tomography (CT) images, which combines an automatic segmentation of the tumor and a DL-based method for extracting rotational-invariant features. Methods: In the first stage, the tumor is segmented from the CT image of the lungs. Here, we use a deep-learning-based method that entails a variational autoencoder to provide more information to a U-Net segmentation model. Next, the 3D volumetric image of the tumor is projected onto 2D spherical maps. These spherical maps serve as inputs for a spherical convolutional neural network that approximates the log risk for a generalized Cox proportional hazard model. Results: The proposed method is compared with 17 baseline methods that combine different feature sets and prediction models using three publicly-available datasets: Lung1 (n=422), Lung3 (n=89), and H&N1 (n=136). We observed comparable C-index scores compared to the best-performing baseline methods in a 5-fold cross-validation on Lung1 (0.59 ± 0.03 vs. 0.62 ± 0.04). In comparison, it slightly outperforms all methods in inter-data set evaluation (0.64 vs. 0.63). The best-performing method from the first experiment reduced its performance to 0.61 and 0.62 for Lung3 and H&N1, respectively. Discussion: The experiments suggest that the performance of spherical features is comparable with previous approaches, but they generalize better when applied to unseen datasets. That might imply that orientation-independent shape features are relevant for SRP. The performance of the proposed method was very similar, using manual and automatic segmentation methods. This makes the proposed model useful in cases where expert annotations are not available or difficult to obtain.

MRI-derived brain age as a biomarker of ageing in rats: validation using a healthy lifestyle intervention.

The difference between brain age predicted from MRI and chronological age (the so-called BrainAGE) has been proposed as an ageing biomarker. We analyse its cross-species potential by testing it on rats undergoing an ageing modulation intervention. Our rat brain age prediction model combined Gaussian process regression with a classifier and achieved a mean absolute error (MAE) of 4.87 weeks using cross-validation on a longitudinal dataset of 31 normal ageing rats. It was then tested on two groups of 24 rats (MAE = 9.89 weeks, correlation coefficient = 0.86): controls vs. a group under long-term environmental enrichment and dietary restriction (EEDR). Using a linear mixed-effects model, BrainAGE was found to increase more slowly with chronological age in EEDR rats (p=0.015 for the interaction term). Cox regression showed that older BrainAGE at 5 months was associated with higher mortality risk (p=0.03). Our findings suggest that lifestyle-related prevention approaches may help to slow down brain ageing in rodents and the potential of BrainAGE as a predictor of age-related health outcomes.

Benign-malignant pulmonary nodule classification in low-dose CT with convolutional features.

PURPOSE: Low-Dose Computed Tomography (LDCT) is the most common imaging modality for lung cancer diagnosis. The presence of nodules in the scans does not necessarily portend lung cancer, as there is an intricate relationship between nodule characteristics and lung cancer. Therefore, benign-malignant pulmonary nodule classification at early detection is a crucial step to improve diagnosis and prolong patient survival. The aim of this study is to propose a method for predicting nodule malignancy based on deep abstract features. METHODS: To efficiently capture both intra-nodule heterogeneities and contextual information of the pulmonary nodules, a dual pathway model was developed to integrate the intra-nodule characteristics with contextual attributes. The proposed approach was implemented with both supervised and unsupervised learning schemes. A random forest model was added as a second component on top of the networks to generate the classification results. The discrimination power of the model was evaluated by calculating the Area Under the Receiver Operating Characteristic Curve (AUROC) metric. RESULTS: Experiments on 1297 manually segmented nodules show that the integration of context and target supervised deep features have a great potential for accurate prediction, resulting in a discrimination power of 0.936 in terms of AUROC, which outperformed the classification performance of the Kaggle 2017 challenge winner. CONCLUSION: Empirical results demonstrate that integrating nodule target and context images into a unified network improves the discrimination power, outperforming the conventional single pathway convolutional neural networks.

A Comparative Study of Radiomics and Deep-Learning Based Methods for Pulmonary Nodule Malignancy Prediction in Low Dose CT Images.

OBJECTIVES: Both radiomics and deep learning methods have shown great promise in predicting lesion malignancy in various image-based oncology studies. However, it is still unclear which method to choose for a specific clinical problem given the access to the same amount of training data. In this study, we try to compare the performance of a series of carefully selected conventional radiomics methods, end-to-end deep learning models, and deep-feature based radiomics pipelines for pulmonary nodule malignancy prediction on an open database that consists of 1297 manually delineated lung nodules. METHODS: Conventional radiomics analysis was conducted by extracting standard handcrafted features from target nodule images. Several end-to-end deep classifier networks, including VGG, ResNet, DenseNet, and EfficientNet were employed to identify lung nodule malignancy as well. In addition to the baseline implementations, we also investigated the importance of feature selection and class balancing, as well as separating the features learned in the nodule target region and the background/context region. By pooling the radiomics and deep features together in a hybrid feature set, we investigated the compatibility of these two sets with respect to malignancy prediction. RESULTS: The best baseline conventional radiomics model, deep learning model, and deep-feature based radiomics model achieved AUROC values (mean ± standard deviations) of 0.792 ± 0.025, 0.801 ± 0.018, and 0.817 ± 0.032, respectively through 5-fold cross-validation analyses. However, after trying out several optimization techniques, such as feature selection and data balancing, as well as adding context features, the corresponding best radiomics, end-to-end deep learning, and deep-feature based models achieved AUROC values of 0.921 ± 0.010, 0.824 ± 0.021, and 0.936 ± 0.011, respectively. We achieved the best prediction accuracy from the hybrid feature set (AUROC: 0.938 ± 0.010). CONCLUSION: The end-to-end deep-learning model outperforms conventional radiomics out of the box without much fine-tuning. On the other hand, fine-tuning the models lead to significant improvements in the prediction performance where the conventional and deep-feature based radiomics models achieved comparable results. The hybrid radiomics method seems to be the most promising model for lung nodule malignancy prediction in this comparative study.

A comparative study of trabecular bone micro-structural measurements using different CT modalities.

Osteoporosis, characterized by reduced bone mineral density and micro-architectural degeneration, significantly enhances fracture-risk. There are several viable methods for trabecular bone micro-imaging, which widely vary in terms of technology, reconstruction principle, spatial resolution, and acquisition time. We have performed an excised cadaveric bone specimen study to evaluate different computed tomography (CT)-imaging modalities for trabecular bone micro-structural analysis. Excised cadaveric bone specimens from the distal radius were scanned using micro-CT and fourin vivoCT imaging modalities: high-resolution peripheral quantitative computed tomography (HR-pQCT), dental cone beam CT (CBCT), whole-body multi-row detector CT (MDCT), and extremity CBCT. A new algorithm was developed to optimize soft thresholding parameters for individualin vivoCT modalities for computing quantitative bone volume fraction maps. Finally, agreement of trabecular bone micro-structural measures, derived from differentin vivoCT imaging, with reference measures from micro-CT imaging was examined. Observed values of most trabecular measures, including trabecular bone volume, network area, transverse and plate-rod micro-structure, thickness, and spacing, forin vivoCT modalities were higher than their micro-CT-based reference values. In general, HR-pQCT-based trabecular bone measures were closer to their reference values as compared to otherin vivoCT modalities. Despite large differences in observed values of measures among modalities, high linear correlation (rε [0.94 0.99]) was found between micro-CT andin vivoCT-derived measures of trabecular bone volume, transverse and plate micro-structural volume, and network area. All HR-pQCT-derived trabecular measures, except the erosion index, showed high correlation (rε [0.91 0.99]). The plate-width measure showed a higher correlation (rε [0.72 0.91]) amongin vivoand micro-CT modalities than its counterpart binary plate-rod characterization-based measure erosion index (rε [0.65 0.81]). Although a strong correlation was observed between micro-structural measures fromin vivoand micro-CT imaging, large shifts in their values forin vivomodalities warrant proper scanner calibration prior to adopting in multi-site and longitudinal studies.

Fully bayesian longitudinal unsupervised learning for the assessment and visualization of AD heterogeneity and progression.

Tau pathology and brain atrophy are the closest correlate of cognitive decline in Alzheimer's disease (AD). Understanding heterogeneity and longitudinal progression of atrophy during the disease course will play a key role in understanding AD pathogenesis. We propose a framework for longitudinal clustering that simultaneously: 1) incorporates whole brain data, 2) leverages unequal visits per individual, 3) compares clusters with a control group, 4) allows for study confounding effects, 5) provides cluster visualization, 6) measures clustering uncertainty. We used amyloid-ß positive AD and negative healthy subjects, three longitudinal structural magnetic resonance imaging scans (cortical thickness and subcortical volume) over two years. We found three distinct longitudinal AD brain atrophy patterns: one typical diffuse pattern (n=34, 47.2%), and two atypical patterns: minimal atrophy (n=23 31.9%) and hippocampal sparing (n=9, 12.5%). We also identified outliers (n=3, 4.2%) and observations with uncertain classification (n=3, 4.2%). The clusters differed not only in regional distributions of atrophy at baseline, but also longitudinal atrophy progression, age at AD onset, and cognitive decline. A framework for the longitudinal assessment of variability in cohorts with several neuroimaging measures was successfully developed. We believe this framework may aid in disentangling distinct subtypes of AD from disease staging.

Shape Information Improves the Cross-Cohort Performance of Deep Learning-Based Segmentation of the Hippocampus.

Performing an accurate segmentation of the hippocampus from brain magnetic resonance images is a crucial task in neuroimaging research, since its structural integrity is strongly related to several neurodegenerative disorders, including Alzheimer's disease (AD). Some automatic segmentation tools are already being used, but, in recent years, new deep learning (DL)-based methods have been proven to be much more accurate in various medical image segmentation tasks. In this work, we propose a DL-based hippocampus segmentation framework that embeds statistical shape of the hippocampus as context information into the deep neural network (DNN). The inclusion of shape information is achieved with three main steps: (1) a U-Net-based segmentation, (2) a shape model estimation, and (3) a second U-Net-based segmentation which uses both the original input data and the fitted shape model. The trained DL architectures were tested on image data of three diagnostic groups [AD patients, subjects with mild cognitive impairment (MCI) and controls] from two cohorts (ADNI and AddNeuroMed). Both intra-cohort validation and cross-cohort validation were performed and compared with the conventional U-net architecture and some variations with other types of context information (i.e., autocontext and tissue-class context). Our results suggest that adding shape information can improve the segmentation accuracy in cross-cohort validation, i.e., when DNNs are trained on one cohort and applied to another. However, no significant benefit is observed in intra-cohort validation, i.e., training and testing DNNs on images from the same cohort. Moreover, compared to other types of context information, the use of shape context was shown to be the most successful in increasing the accuracy, while keeping the computational time in the order of a few minutes.

Assessment of image quality in abdominal computed tomography: Effect of model-based iterative reconstruction, multi-planar reconstruction and slice thickness on potential dose reduction.

PURPOSE: To determine the effect of tube load, model-based iterative reconstruction (MBIR) strength and slice thickness in abdominal CT using visual comparison of multi-planar reconstruction images. METHOD: Five image criteria were assessed independently by four radiologists on two data sets at 42- and 98-mAs tube loads for 25 patients examined on a 192-slice dual-source CT scanner. Effect of tube load, MBIR strength, slice thickness and potential dose reduction was estimated with Visual Grading Regression (VGR). Objective image quality was determined by measuring noise (SD), contrast-to-noise (CNR) ratio and noise-power spectra (NPS). RESULTS: Comparing 42- and 98-mAs tube loads, improved image quality was observed as a strong effect of log tube load regardless of MBIR strength (p < 0.001). Comparing strength 5 to 3, better image quality was obtained for two criteria (p < 0.01), but inferior for liver parenchyma and overall image quality. Image quality was significantly better for slice thicknesses of 2mm and 3mm compared to 1mm, with potential dose reductions between 24%-41%. As expected, with decrease in slice thickness and algorithm strength, the noise power and SD (HU-values) increased, while the CNR decreased. CONCLUSION: Increasing slice thickness from 1 mm to 2 mm or 3 mm allows for a possible dose reduction. MBIR strength 5 shows improved image quality for three out of five criteria for 1 mm slice thickness. Increasing MBIR strength from 3 to 5 has diverse effects on image quality. Our findings do not support a general recommendation to replace strength 3 by strength 5 in clinical abdominal CT protocols. However, strength 5 may be used in task-based protocols.