Artificial Intelligence-Based Prediction of Contrast Medium Doses for Computed Tomography Angiography Using Optimized Clinical Parameter Sets.

OBJECTIVES: In this paper, an artificial intelligence-based algorithm for predicting the optimal contrast medium dose for computed tomography (CT) angiography of the aorta is presented and evaluated in a clinical study. The prediction of the contrast dose reduction is modelled as a classification problem using the image contrast as the main feature. METHODS: This classification is performed by random decision forests (RDF) and k-nearest-neighbor methods (KNN). For the selection of optimal parameter subsets all possible combinations of the 22 clinical parameters (age, blood pressure, etc.) are considered using the classification accuracy and precision of the KNN classifier and RDF as quality criteria. Subsequently, the results of the evaluation were optimized by means of feature transformation using regression neural networks (RNN). These were used for a direct classification based on regressed Hounsfield units as well as preprocessing for a subsequent KNN classification. RESULTS: For feature selection, an RDF model achieved the highest accuracy of 84.42% and a KNN model achieved the best precision of 86.21%. The most important parameters include age, height, and hemoglobin. The feature transformation using an RNN considerably exceeded these values with an accuracy of 90.00% and a precision of 97.62% using all 22 parameters as input. However, also the feasibility of the parameter sets in routine clinical practice has to be considered, because some of the 22 parameters are not measured in routine clinical practice and additional measurement time of 15 to 20 minutes per patient is needed. Using the standard feature set available in clinical routine the best accuracy of 86.67% and precision of 93.18% was achieved by the RNN. CONCLUSION: We developed a reliable hybrid system that helps radiologists determine the optimal contrast dose for CT angiography based on patient-specific parameters.

Random Forest and Gradient Boosted Trees for Patient Individualized Contrast Agent Dose Reduction in CT Angiography.

This work aims to recognize the patient individual possibility of contrast dose reduction in CT angiography. This system should help to identify whether the dose of contrast agent in CT angiography can be reduced to avoid side effects. In a clinical study, 263 CT angiographies were performed and, in addition, 21 clinical parameters were recorded for each patient before contrast agent administration. The resulting images were labeled according to their contrast quality. It is assumed that the contrast dose could be reduced for CT angiography images with excessive contrast. These data was used to develop a model for predicting excessive contrast based on the clinical parameters using logistic regression, random forest, and gradient boosted trees. In addition, the minimization of clinical parameters required was investigated to reduce the overall effort. Therefore, models were tested with all subsets of clinical parameters and each parameter's importance was examined. In predicting excessive contrast in CT angiography images covering the aortic region, a maximum accuracy of 0.84 was achieved by a random forest with 11 clinical parameters; for the leg-pelvis region data, an accuracy of 0.87 was achieved by a random forest with 7 parameters; and for the entire data set, an accuracy of 0.74 was achieved by gradient boosted trees with 9 parameters.

Automated Motor Tic Detection: A Machine Learning Approach.

BACKGROUND: Video-based tic detection and scoring is useful to independently and objectively assess tic frequency and severity in patients with Tourette syndrome. In trained raters, interrater reliability is good. However, video ratings are time-consuming and cumbersome, particularly in large-scale studies. Therefore, we developed two machine learning (ML) algorithms for automatic tic detection. OBJECTIVE: The aim of this study was to evaluate the performances of state-of-the-art ML approaches for automatic video-based tic detection in patients with Tourette syndrome. METHODS: We used 64 videos of n = 35 patients with Tourette syndrome. The data of six subjects (15 videos with ratings) were used as a validation set for hyperparameter optimization. For the binary classification task to distinguish between tic and no-tic segments, we established two different supervised learning approaches. First, we manually extracted features based on landmarks, which served as input for a Random Forest classifier (Random Forest). Second, a fully automated deep learning approach was used, where regions of interest in video snippets were input to a convolutional neural network (deep neural network). RESULTS: Tic detection F1 scores (and accuracy) were 82.0% (88.4%) in the Random Forest and 79.5% (88.5%) in the deep neural network approach. CONCLUSIONS: ML algorithms for automatic tic detection based on video recordings are feasible and reliable and could thus become a valuable assessment tool, for example, for objective tic measurements in clinical trials. ML algorithms might also be useful for the differential diagnosis of tics. © 2023 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Chaos analysis of the cortical boundary for the recognition of psychosis.

BACKGROUND: Structural MRI studies in people with first-episode psychosis (FEP) and those in the clinical high-risk (CHR) state have consistently shown volumetric abnormalities that depict changes in the structural complexity of the cortical boundary. The aim of the present study was to employ chaos analysis in the identification of people with psychosis based on the structural complexity of the cortical boundary and subcortical areas. METHODS: We performed chaos analysis of the grey matter distribution on structural MRIs. First, the outer boundary points for each slice in the axial, coronal and sagittal view were calculated for grey matter maps. Next, the distance of each boundary point from the centre of mass in the grey matter was calculated and stored as spatial series, which was further analyzed by extracting the Largest Lyapunov Exponent (lambda [λ]), a feature depicting the structural complexity of the cortical boundary. RESULTS: Structural MRIs were acquired from 77 FEP, 73 CHR and 44 healthy controls. We compared λ brain maps between groups, which resulted in statistically significant differences in all comparisons. By matching the λ values extracted in axial view with the Morlet wavelet, differences on the surface relief are observed between groups. LIMITATIONS: Parameters were selected after experimentation on the examined sample. Investigation of the effectiveness of the method in a larger data set is needed. CONCLUSION: The proposed framework using spatial series verifies diagnosis-relevant features and may contribute to the identification of structural biomarkers for psychosis.

Chaos analysis of the brain topology in first-episode psychosis and clinical high risk patients.

Structural MRI studies in first-episode psychosis (FEP) and in clinical high risk (CHR) patients have consistently shown volumetric abnormalities in frontal, temporal, and cingulate cortex areas. The aim of the present study was to employ chaos analysis for the identification of brain topology differences in people with psychosis. Structural MRI were acquired from 77 FEP, 73 CHR and 44 healthy controls (HC). Chaos analysis of the gray matter distribution was performed: First, the distances of each voxel from the center of mass in the gray matter image was calculated. Next, the distances multiplied by the voxel intensity were represented as a spatial-series, which then was analyzed by extracting the Largest-Lyapunov-Exponent (lambda). The lambda brain map depicts thus how the gray matter topology changes. Between-group differences were identified by (a) comparing the lambda brain maps, which resulted in statistically significant differences in FEP and CHR compared to HC; and (b) matching the lambda series with the Morlet wavelet, which resulted in statistically significant differences in the scalograms of FEP against CHR and HC. The proposed framework using spatial-series extraction enhances the between-group differences of FEP, CHR and HC subjects, verifies diagnosis-relevant features and may potentially contribute to the identification of structural biomarkers for psychosis.

Differentiation of different stages of brain tumor infiltration using optical coherence tomography: Comparison of two systems and histology.

The discrimination of tumor-infiltrated tissue from non-tumorous brain tissue during neurosurgical tumor excision is a major challenge in neurosurgery. It is critical to achieve full tumor removal since it directly correlates with the survival rate of the patient. Optical coherence tomography (OCT) might be an additional imaging method in the field of neurosurgery that enables the classification of different levels of tumor infiltration and non-tumorous tissue. This work investigated two OCT systems with different imaging wavelengths (930 nm/1310 nm) and different resolutions (axial (air): 4.9 µm/16 µm, lateral: 5.2 µm/22 µm) in their ability to identify different levels of tumor infiltration based on freshly excised ex vivo brain samples. A convolutional neural network was used for the classification. For both systems, the neural network could achieve classification accuracies above 91% for discriminating between healthy white matter and highly tumor infiltrated white matter (tumor infiltration >60%) .This work shows that both OCT systems with different optical properties achieve similar results regarding the identification of different stages of brain tumor infiltration.

Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection.

Manual detection of newly formed lesions in multiple sclerosis is an important but tedious and difficult task. Several approaches for automating the detection of new lesions have recently been proposed, but they tend to either overestimate the actual amount of new lesions or to miss many lesions. In this paper, an image registration convolutional neural network (CNN) that adapts the baseline image to the follow-up image by spatial deformations and simulation of new lesions is proposed. Simultaneously, segmentations of new lesions are generated, which are shown to reliably estimate the real new lesion load and to separate stable and progressive patients. Several applications of the proposed network emerge: image registration, detection and segmentation of new lesions, and modeling of new MS lesions. The modeled lesions offer the possibility to investigate the intensity profile of new lesions.

Chronological Registration of OCT and Autofluorescence Findings in CSCR: Two Distinct Patterns in Disease Course.

Optical coherence tomography (OCT) and fundus autofluorescence (FAF) are important imaging modalities for the assessment and prognosis of central serous chorioretinopathy (CSCR). However, setting the findings from both into spatial and temporal contexts as desirable for disease analysis remains a challenge due to both modalities being captured in different perspectives: sparse three-dimensional (3D) cross sections for OCT and two-dimensional (2D) en face images for FAF. To bridge this gap, we propose a visualisation pipeline capable of projecting OCT labels to en face image modalities such as FAF. By mapping OCT B-scans onto the accompanying en face infrared (IR) image and then registering the IR image onto the FAF image by a neural network, we can directly compare OCT labels to other labels in the en face plane. We also present a U-Net inspired segmentation model to predict segmentations in unlabeled OCTs. Evaluations show that both our networks achieve high precision (0.853 Dice score and 0.913 Area under Curve). Furthermore, medical analysis performed on exemplary, chronologically arranged CSCR progressions of 12 patients visualized with our pipeline indicates that, on CSCR, two patterns emerge: subretinal fluid (SRF) in OCT preceding hyperfluorescence (HF) in FAF and vice versa.

Proposal of Semantic Annotation for German Metadata Using Bidirectional Recurrent Neural Networks.

The distributed nature of our digital healthcare and the rapid emergence of new data sources prevents a compelling overview and the joint use of new data. Data integration, e.g., with metadata and semantic annotations, is expected to overcome this challenge. In this paper, we present an approach to predict UMLS codes to given German metadata using recurrent neural networks. The augmentation of the training dataset using the Medical Subject Headings (MeSH), particularly the German translations, also improved the model accuracy. The model demonstrates robust performance with 75% accuracy and aims to show that increasingly sophisticated machine learning tools can already play a significant role in data integration.

Registration of histological brain images onto optical coherence tomography images based on shape information.

Identifying tumour infiltration zones during tumour resection in order to excise as much tumour tissue as possible without damaging healthy brain tissue is still a major challenge in neurosurgery. The detection of tumour infiltrated regions so far requires histological analysis of biopsies taken from at expected tumour boundaries. The gold standard for histological analysis is the staining of thin cut specimen and the evaluation by a neuropathologist. This work presents a way to transfer the histological evaluation of a neuropathologist onto optical coherence tomography (OCT) images. OCT is a method suitable for real timein vivoimaging during neurosurgery however the images require processing for the tumour detection. The method demonstrated here enables the creation of a dataset which will be used for supervised learning in order to provide a better visualization of tumour infiltrated areas for the neurosurgeon. The created dataset contains labelled OCT images from two different OCT-systems (wavelength of 930 nm and 1300 nm). OCT images corresponding to the stained histological images were determined by shaping the sample, a controlled cutting process and a rigid transformation process between the OCT volumes based on their topological information. The histological labels were transferred onto the corresponding OCT images through a non-rigid transformation based on shape context features retrieved from the sample outline in the histological image and the OCT image. The accuracy of the registration was determined to be 200 ± 120µm. The resulting dataset consists of 1248 labelled OCT images for each of the two OCT systems.

Deep learning-based simultaneous registration and unsupervised non-correspondence segmentation of medical images with pathologies.

PURPOSE: The registration of medical images often suffers from missing correspondences due to inter-patient variations, pathologies and their progression leading to implausible deformations that cause misregistrations and might eliminate valuable information. Detecting non-corresponding regions simultaneously with the registration process helps generating better deformations and has been investigated thoroughly with classical iterative frameworks but rarely with deep learning-based methods. METHODS: We present the joint non-correspondence segmentation and image registration network (NCR-Net), a convolutional neural network (CNN) trained on a Mumford-Shah-like functional, transferring the classical approach to the field of deep learning. NCR-Net consists of one encoding and two decoding parts allowing the network to simultaneously generate diffeomorphic deformations and segment non-correspondences. The loss function is composed of a masked image distance measure and regularization of deformation field and segmentation output. Additionally, anatomical labels are used for weak supervision of the registration task. No manual segmentations of non-correspondences are required. RESULTS: The proposed network is evaluated on the publicly available LPBA40 dataset with artificially added stroke lesions and a longitudinal optical coherence tomography (OCT) dataset of patients with age-related macular degeneration. The LPBA40 data are used to quantitatively assess the segmentation performance of the network, and it is shown qualitatively that NCR-Net can be used for the unsupervised segmentation of pathologies in OCT images. Furthermore, NCR-Net is compared to a registration-only network and state-of-the-art registration algorithms showing that NCR-Net achieves competitive performance and superior robustness to non-correspondences. CONCLUSION: NCR-Net, a CNN for simultaneous image registration and unsupervised non-correspondence segmentation, is presented. Experimental results show the network's ability to segment non-correspondence regions in an unsupervised manner and its robust registration performance even in the presence of large pathologies.

A systematic comparison of generative models for medical images.

PURPOSE: This work aims for a systematic comparison of popular shape and appearance models. Here, two statistical and four deep-learning-based shape and appearance models are compared and evaluated in terms of their expressiveness described by their generalization ability and specificity as well as further properties like input data format, interpretability and latent space distribution and dimension. METHODS: Classical shape models and their locality-based extension are considered next to autoencoders, variational autoencoders, diffeomorphic autoencoders and generative adversarial networks. The approaches are evaluated in terms of generalization ability, specificity and likeness depending on the amount of training data. Furthermore, various latent space metrics are presented in order to capture further major characteristics of the models. RESULTS: The experimental setup showed that locality statistical shape models yield best results in terms of generalization ability for 2D and 3D shape modeling. However, the deep learning approaches show strongly improved specificity. In the case of simultaneous shape and appearance modeling, the neural networks are able to generate more realistic and diverse appearances. A major drawback of the deep-learning models is, however, their impaired interpretability and ambiguity of the latent space. CONCLUSIONS: It can be concluded that for applications not requiring particularly good specificity, shape modeling can be reliably established with locality-based statistical shape models, especially when it comes to 3D shapes. However, deep learning approaches are more worthwhile in terms of appearance modeling.

Memory-efficient GAN-based domain translation of high resolution 3D medical images.

Generative adversarial networks (GANs) are currently rarely applied on 3D medical images of large size, due to their immense computational demand. The present work proposes a multi-scale patch-based GAN approach for establishing unpaired domain translation by generating 3D medical image volumes of high resolution in a memory-efficient way. The key idea to enable memory-efficient image generation is to first generate a low-resolution version of the image followed by the generation of patches of constant sizes but successively growing resolutions. To avoid patch artifacts and incorporate global information, the patch generation is conditioned on patches from previous resolution scales. Those multi-scale GANs are trained to generate realistically looking images from image sketches in order to perform an unpaired domain translation. This allows to preserve the topology of the test data and generate the appearance of the training domain data. The evaluation of the domain translation scenarios is performed on brain MRIs of size 155 × 240 × 240 and thorax CTs of size up to 5123. Compared to common patch-based approaches, the multi-resolution scheme enables better image quality and prevents patch artifacts. Also, it ensures constant GPU memory demand independent from the image size, allowing for the generation of arbitrarily large images.

Automatic quality measurement of aortic contrast-enhanced CT angiographies for patient-specific dose optimization.

PURPOSE: Iodine-containing contrast agent (CA) used in contrast-enhanced CT angiography (CTA) can pose a health risk for patients. A system that adjusts the frequently used standard CA dose for individual patients based on their clinical parameters can be useful. As basis the quality of the image contrast in CTA volumes has to be determined, especially to recognize excessive contrast induced by CA overdosing. However, a manual assessment with a ROI-based image contrast classification is a time-consuming step in everyday clinical practice. METHODS: We propose a method to automate the contrast measurement of aortic CTA volumes. The proposed algorithm is based on the mean HU values in selected ROIs that were automatically positioned in the CTA volume. First, an automatic localization algorithm determines the CTA image slices for certain ROIs followed by the localization of these ROIs. A rule-based classification using the mean HU values in the ROIs categorizes images with insufficient, optimal and excessive contrast. RESULTS: In 95.89% (70 out of 73 CTAs obtained with the ulrich medical CT motion contrast media injector) the algorithm chose the same image contrast class as the radiological expert. The critical case of missing an overdose did not occur with a positive predicative value of 100%. CONCLUSION: The resulting system works well within our range of considered scan protocols detecting enhanced areas in CTA volumes. Our work automized an assessment for classifying CA-induced image contrast which reduces the time needed for medical practitioners to perform such an assessment manually.

Computer-Assisted Quality Assessment of Aortic CT Angiographies for Patient-Individual Dose Adjustment.

Iodine-containing contrast agents (CA) are important for enhanced image contrast in CT imaging especially in CT angiography (CTA). CA however poses a risk to the patient since it can e.g. harm the kidneys. In clinical routine often a standard dose is applied that does not take differences between individual patients into account. We propose a method that as a preliminary stage determines excessive image contrast and CA overdosing by assessing the image contrast in CTA images obtained with the ulrich medical CT motion contrast media injector with RIS/PACS interface. A resulting CA dose recommendation is linked to a set of clinical parameters collected for each assessed patient. We used the established data set to implement an automatic classification for individual CA dose adjustment. The classification determines similar cases of new patients to take on the associated CA dose adjustment recommendation. The computation of similar patient data is based on the previously collected patient-individual parameters. The study shows that as basis for a recommendations the largest proportion of patients receive too much CA. A first evaluation of the automatic classification showed an overall error rate of 22% to recognize the correct class for CA dose adjustments using a k-NN-Classifier and a leave-one-out method. The classification's positive predictive value for correctly assigning a CA overdosing was 85.71%.

Evaluating White Matter Lesion Segmentations with Refined Sørensen-Dice Analysis.

The Sørensen-Dice index (SDI) is a widely used measure for evaluating medical image segmentation algorithms. It offers a standardized measure of segmentation accuracy which has proven useful. However, it offers diminishing insight when the number of objects is unknown, such as in white matter lesion segmentation of multiple sclerosis (MS) patients. We present a refinement for finer grained parsing of SDI results in situations where the number of objects is unknown. We explore these ideas with two case studies showing what can be learned from our two presented studies. Our first study explores an inter-rater comparison, showing that smaller lesions cannot be reliably identified. In our second case study, we demonstrate fusing multiple MS lesion segmentation algorithms based on the insights into the algorithms provided by our analysis to generate a segmentation that exhibits improved performance. This work demonstrates the wealth of information that can be learned from refined analysis of medical image segmentations.

Segmentation of mouse skin layers in optical coherence tomography image data using deep convolutional neural networks.

Optical coherence tomography (OCT) enables the non-invasive acquisition of high-resolution three-dimensional cross-sectional images at micrometer scale and is mainly used in the field of ophthalmology for diagnosis as well as monitoring of eye diseases. Also in other areas, such as dermatology, OCT is already well established. Due to its non-invasive nature, OCT is also employed for research studies involving animal models. Manual evaluation of OCT images of animal models is a challenging task due to the lack of imaging standards and the varying anatomy among models. In this paper, we present a deep learning algorithm for the automatic segmentation of several layers of mouse skin in OCT image data using a deep convolutional neural network (CNN). The architecture of our CNN is based on the U-net and is modified by densely connected convolutions. We compared our adapted CNN with our previous algorithm, a combination of a random forest classification and a graph-based refinement, and a baseline U-net. The results showed that, on average, our proposed CNN outperformed our previous algorithm and the baseline U-net. In addition, a reduction of outliers could be observed through the use of densely connected convolutions.

Robust GPU-based virtual reality simulation of radio-frequency ablations for various needle geometries and locations.

PURPOSE: Radio-frequency ablations play an important role in the therapy of malignant liver lesions. The navigation of a needle to the lesion poses a challenge for both the trainees and intervening physicians. METHODS: This publication presents a new GPU-based, accurate method for the simulation of radio-frequency ablations for lesions at the needle tip in general and for an existing visuo-haptic 4D VR simulator. The method is implemented real time capable with Nvidia CUDA. RESULTS: It performs better than a literature method concerning the theoretical characteristic of monotonic convergence of the bioheat PDE and a in vitro gold standard with significant improvements ([Formula: see text]) in terms of Pearson correlations. It shows no failure modes or theoretically inconsistent individual simulation results after the initial phase of 10 s. On the Nvidia 1080 Ti GPU, it achieves a very high frame rendering performance of > 480 Hz. CONCLUSION: Our method provides a more robust and safer real-time ablation planning and intraoperative guidance technique, especially avoiding the overestimation of the ablated tissue death zone, which is risky for the patient in terms of tumor recurrence. Future in vitro measurements and optimization shall further improve the conservative estimate.