Fast and accurate 3-D spine MRI segmentation using FastCleverSeg

Accurate and efficient segmenting of vertebral bodies, muscles, and discs is crucial for analyzing various spinal diseases. However, traditional methods are either laborious and time-consuming (manual segmentation) or require extensive training data (fully automatic segmentation). FastCleverSeg, our proposed semi-automatic segmentation approach, addresses those limitations by significantly reducing user interaction while maintaining high accuracy. First, we reduce user interaction by requiring the manual annotation of only two or three slices. Next, we automatically Estimate the Annotation on Intermediary Slices (EANIS) using traditional computer vision/graphics concepts. Finally, our proposed method leverages improved voxel weight balancing to achieve fast and precise volumetric segmentation in the segmentation process. Experimental evaluations on our assembled diverse MRI databases comprising 179 patients (60 male, 119 female), demonstrate a remarkable 25 ms (30 ms standard deviation) processing time and a significant reduction in user interaction compared to existing approaches. Importantly, FastCleverSeg maintains or surpasses the segmentation quality of competing methods, achieving a Dice score of 94%. This invaluable tool empowers physicians to efficiently generate reliable ground truths, expediting the segmentation process and paving the way for future integration with deep learning approaches. In turn, this opens exciting possibilities for future fully automated spine segmentation.

A 3D Radiomics-Based Artificial Neural Network Model for Benign Versus Malignant Vertebral Compression Fracture Classification in MRI

To train an artificial neural network model using 3D radiomic features to differentiate benign from malignant vertebral compression fractures (VCFs) on MRI. This retrospective study analyzed sagittal T1-weighted lumbar spine MRIs from 91 patients (average age of 64.24 ± 11.75 years) diagnosed with benign or malignant VCFs from 2010 to 2019, of them 47 (51.6%) had benign VCFs and 44 (48.4%) had malignant VCFs. The lumbar fractures were three-dimensionally segmented and had their radiomic features extracted and selected with the wrapper method. The training set consisted of 100 fractured vertebral bodies from 61 patients (average age of 63.2 ± 12.5 years), and the test set was comprised of 30 fractured vertebral bodies from 30 patients (average age of 66.4 ± 9.9 years). Classification was performed with the multilayer perceptron neural network with a back-propagation algorithm. To validate the model, the tenfold cross-validation technique and an independent test set (holdout) were used. The performance of the model was evaluated using the average with a 95% confidence interval for the ROC AUC, accuracy, sensitivity, and specificity (considering the threshold = 0.5). In the internal validation test, the best model reached a ROC AUC of 0.98, an accuracy of 95% (95/100), a sensitivity of 93.5% (43/46), and specificity of 96.3% (52/54). In the validation with independent test set, the model achieved a ROC AUC of 0.97, an accuracy of 93.3% (28/30), a sensitivity of 93.3% (14/15), and a specificity of 93.3% (14/15). The model proposed in this study using radiomic features could differentiate benign from malignant vertebral compression fractures with excellent performance and is promising as an aid to radiologists in the characterization of VCFs.

Noise modeling and variance stabilization of a computed radiography (CR) mammography system subject to fixed-pattern noise

In this work we model the noise properties of a computed radiography (CR) mammography system by adding an extra degree of freedom to a well-established noise model, and derive a variance-stabilizing transform (VST) to convert the signal-dependent noise into approximately signal-independent. The proposed model relies on a quadratic variance function, which considers fixed-pattern (structural), quantum and electronic noise. It also accounts for the spatial-dependency of the noise by assuming a space-variant quantum coefficient. The proposed noise model was compared against two alternative models commonly found in the literature. The first alternative model ignores the spatial-variability of the quantum noise, and the second model assumes negligible structural noise. We also derive a VST to convert noisy observations contaminated by the proposed noise model into observations with approximately Gaussian noise and constant variance equals to one. Finally, we estimated a look-up table that can be used as an inverse transform in denoising applications. A phantom study was conducted to validate the noise model, VST and inverse VST. The results show that the space-variant signal-dependent quadratic noise model is appropriate to describe noise in this CR mammography system (errors< 2.0% in terms of signal-to-noise ratio). The two alternative noise models were outperformed by the proposed model (errors as high as 14.7% and 9.4%). The designed VST was able to stabilize the noise so that it has variance approximately equal to one (errors< 4.1%), while the two alternative models achieved errors as high as 26.9% and 18.0%, respectively. Finally, the proposed inverse transform was capable of returning the signal to the original signal range with virtually no bias.

Manual and semiautomatic segmentation of bone sarcomas on MRI have high similarity

The aims of this study were to evaluate the intra- and interobserver reproducibility of manual segmentation of bone sarcomas in magnetic resonance imaging (MRI) studies and to compare manual and semiautomatic segmentation methods. This retrospective study included twelve osteosarcoma and eight Ewing sarcoma MRI studies performed prior to any therapeutic intervention. All cases were histopathologically confirmed. Three radiologists used 3D-Slicer software to perform manual segmentation of bone sarcomas in a blinded and independent manner. One radiologist segmented manually and also performed semiautomatic segmentation with the GrowCut tool. Segmentation exercises were timed for comparison. The dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to evaluate similarity between the segmentation results and further statistical analyses were performed to compare DSC, HD, and volumetric results. Manual segmentation was reproducible with intraobserver DSC varying from 0.83 to 0.97 and HD from 3.37 to 28.73 mm. Interobserver DSC of manual segmentation showed variation from 0.73 to 0.97 and HD from 3.93 to 33.40 mm. Semiautomatic segmentation compared to manual segmentation resulted in DSCs of 0.71−0.96 and HDs of 5.38−31.54 mm. Semiautomatic segmentation required significantly less time compared to manual segmentation (P value ≤0.05). Among all situations compared, tumor volumetry did not show significant statistical differences (P value >0.05). We found excellent intra- and interobserver agreement for manual segmentation of osteosarcoma and Ewing sarcoma. There was high similarity between manual and semiautomatic segmentation, with a significant reduction of segmentation time using the semiautomatic method.

A superpixel-driven deep learning approach for the analysis of dermatological wounds

BACKGROUND

The image-based identification of distinct tissues within dermatological wounds enhances patients' care since it requires no intrusive evaluations. This manuscript presents an approach, we named QTDU, that combines deep learning models with superpixel-driven segmentation methods for assessing the quality of tissues from dermatological ulcers.

METHOD

QTDU consists of a three-stage pipeline for the obtaining of ulcer segmentation, tissues' labeling, and wounded area quantification. We set up our approach by using a real and annotated set of dermatological ulcers for training several deep learning models to the identification of ulcered superpixels.

RESULTS

Empirical evaluations on 179,572 superpixels divided into four classes showed QTDU accurately spot wounded tissues (AUC = 0.986, sensitivity = 0.97, and specificity = 0.974) and outperformed machine-learning approaches in up to 8.2% regarding F1-Score through fine-tuning of a ResNet-based model. Last, but not least, experimental evaluations also showed QTDU correctly quantified wounded tissue areas within a 0.089 Mean Absolute Error ratio.

CONCLUSIONS

Results indicate QTDU effectiveness for both tissue segmentation and wounded area quantification tasks. When compared to existing machine-learning approaches, the combination of superpixels and deep learning models outperformed the competitors within strong significant levels.

Towards an Information Infrastructure for Medical Image Sharing

Information infrastructures involve the notion of a shared, open infrastructure, constituting a space where people, organizations, and technical components associate to develop an activity. The current infrastructure for medical image sharing, based on PACS/DICOM technologies, does not constitute an information infrastructure since it is limited in its ability to share in a scalable, comprehensive, and secure manner. This paper proposes the DICOMFlow, a decentralized, distributed infrastructure model that aims to foment the formation of an information infrastructure in order to share medical images and teleradiology. As an installed base, it uses the PACS/DICOM infrastructure of radiology departments and the internet e-mail infrastructure. Experiments performed in real and simulated environments have indicated the feasibility of the proposed infrastructure to foment the formation of an information infrastructure for medical image sharing and teleradiology.

Intraoral radiographs texture analysis for dental implant planning

BACKGROUND AND OBJECTIVES

Computer vision extracts features or attributes from images improving diagnosis accuracy and aiding in clinical decisions. This study aims to investigate the feasibility of using texture analysis of periapical radiograph images as a tool for dental implant treatment planning.

METHODS

Periapical radiograph images of 127 jawbone sites were obtained before and after implant placement. From the superimposition of the pre- and post-implant images, four regions of interest (ROI) were delineated on the pre-implant images for each implant site: mesial, distal and apical peri-implant areas and a central area. Each ROI was analysed using Matlab® software and seven image attributes were extracted: mean grey level (MGL), standard deviation of grey levels (SDGL), coefficient of variation (CV), entropy (En), contrast, correlation (Cor) and angular second moment (ASM). Images were grouped by bone types-Lekholm and Zarb classification (1,2,3,4). Peak insertion torque (PIT) and resonance frequency analysis (RFA) were recorded during implant placement. Differences among groups were tested for each image attribute. Agreement between measurements of the peri-implant ROIs and overall ROI (peri-implant + central area) was tested, as well as the association between primary stability measures (PIT and RFA) and texture attributes.

RESULTS

Differences among bone type groups were found for MGL (p = 0.035), SDGL (p = 0.024), CV (p < 0.001) and En (p < 0.001). The apical ROI showed a significant difference from the other regions for all attributes, except Cor. Concordance correlation coefficients were all almost perfect (ρ > 0.93), except for ASM (ρ = 0.62). Texture attributes were significantly associated with the implant stability measures.

CONCLUSION

Texture analysis of periapical radiographs may be a reliable non-invasive quantitative method for the assessment of jawbone and prediction of implant stability, with potential clinical applications.

Fuzzy membership functions for analysis of high-resolution CT images of diffuse pulmonary diseases

We propose the use of fuzzy membership functions to analyze images of diffuse pulmonary diseases (DPDs) based on fractal and texture features. The features were extracted from preprocessed regions of interest (ROIs) selected from high-resolution computed tomography images. The ROIs represent five different patterns of DPDs and normal lung tissue. A Gaussian mixture model (GMM) was constructed for each feature, with six Gaussians modeling the six patterns. Feature selection was performed and the GMMs of the five significant features were used. From the GMMs, fuzzy membership functions were obtained by a probability-possibility transformation and further statistical analysis was performed. An average classification accuracy of 63.5% was obtained for the six classes. For four of the six classes, the classification accuracy was superior to 65%, and the best classification accuracy was 75.5% for one class. The use of fuzzy membership functions to assist in pattern classification is an alternative to deterministic approaches to explore strategies for medical diagnosis.

Classification of vertebral compression fractures in magnetic resonance images using spectral and fractal analysis

Fractures with partial collapse of vertebral bodies are generically referred to as "vertebral compression fractures" or VCFs. VCFs can have different etiologies comprising trauma, bone failure related to osteoporosis, or metastatic cancer affecting bone. VCFs related to osteoporosis (benign fractures) and to cancer (malignant fractures) are commonly found in the elderly population. In the clinical setting, the differentiation between benign and malignant fractures is complex and difficult. This paper presents a study aimed at developing a system for computer-aided diagnosis to help in the differentiation between malignant and benign VCFs in magnetic resonance imaging (MRI). We used T1-weighted MRI of the lumbar spine in the sagittal plane. Images from 47 consecutive patients (31 women, 16 men, mean age 63 years) were studied, including 19 malignant fractures and 54 benign fractures. Spectral and fractal features were extracted from manually segmented images of 73 vertebral bodies with VCFs. The classification of malignant vs. benign VCFs was performed using the k-nearest neighbor classifier with the Euclidean distance. Results obtained show that combinations of features derived from Fourier and wavelet transforms, together with the fractal dimension, were able to obtain correct classification rate up to 94.7% with area under the receiver operating characteristic curve up to 0.95.

PRoSPer: perceptual similarity queries in medical CBIR systems through user profiles

In this paper, we present a novel approach to perform similarity queries over medical images, maintaining the semantics of a given query posted by the user. Content-based image retrieval systems relying on relevance feedback techniques usually request the users to label relevant/irrelevant images. Thus, we present a highly effective strategy to survey user profiles, taking advantage of such labeling to implicitly gather the user perceptual similarity. The profiles maintain the settings desired for each user, allowing tuning of the similarity assessment, which encompasses the dynamic change of the distance function employed through an interactive process. Experiments on medical images show that the method is effective and can improve the decision making process during analysis.

Detection of vertebral plateaus in lateral lumbar spinal X-ray images with Gabor filters

A few recent studies have proposed computed-aided methods for the detection and analysis of vertebral bodies in radiographic images. This paper presents a method based on Gabor filters. Forty-one lateral lumbar spinal X-ray images from different patients were included in the study. For each image, a radiologist manually delineated the vertebral plateaus of L1, L2, L3, and L4 using a software tool for image display and mark-up. Each original image was filtered with a bank of 180 Gabor filters. The angle of the Gabor filter with the highest response at each pixel was used to derive a measure of the strength of orientation or alignment. In order to limit the spatial extent of the image data and the derived features in further analysis, a semi-automated procedure was applied to the original image. A neural network utilizing the logistic sigmoid function was trained with pixel intensity from the original image, the result of manual delineation of the plateaus, the Gabor magnitude response, and the alignment image. The average overlap between the results of detection by image processing and manual delineation of the plateaus of L1-L4 in the 41 images tested was 0.917. The results are expected to be useful in the analysis of vertebral deformities and fractures.

Medical image registration using TSallis Entropy in Statistical Parametric Mapping (SPM)

The superposition of medical images, technically known as co-registration, can take a major role in determining the topographic and morphological changes in functional diagnostic and therapeutic purposes. This paper describes a study focused on to find an alternative cost function method for medical images co-registration through the study of performance and robustness of the TSallis Entropy in Statistical Parametric Mapping package (SPM). Images of Magnetic Resonance (MR) and Single Photon Emission Computed Tomography (SPECT) of 3 patients morphologically normal were used for the construction of anatomic phantoms containing predetermined geometric variations. The simulated images were co-registered with the original images using traditional techniques and the proposed method. The comparative analysis of the Root Mean Square (RMS) error showed that the Tsallis Entropy was more efficient in the intramodality alignment, while the Shannon Entropy in the intermodality one; revealing therefore the importance of the implementation of the Tsallis Entropy in SPM for applications in neurology and neuropsychiatric evaluation.

Using relevance feedback to reduce the semantic gap in content-based image retrieval of mammographic masses

This paper presents the use of relevance feedback (RFb) to reduce the semantic gap in content-based image retrieval (CBIR) of mammographic masses. Tests were conducted where the radiologists' classification of the lesions based on the BI-RADS categories were used with techniques of query-point movement to incorporate RFb. The measures of similarity of images used for CBIR were based upon Zernike moments. The performance of CBIR was measured in terms of precision and recall of retrieval. The results indicate improvement due to RFb of up to 41.6% in precision. In our experiments, the gain in the performance of CBIR with RFb was associated with the BI-RADS category of the query mammographic image, with large improvement in cases of lesions belonging to categories 4 and 5. The proposed method could find applications in computer-aided diagnosis (CAD) of breast cancer.

Color image processing and content-based image retrieval techniques for the analysis of dermatological lesions

This paper presents color image processing methods for the analysis of dermatological images in the context of a content-based image retrieval (CBIR) system. Tests were conducted on the classification of tissue components in skin lesions, in terms of necrotic tissue, fibrin, granulation, and mixed composition. The images were classified based on color components by an expert dermatologist following a black-yellow-red model. Indexing and retrieval of images were performed based on texture information obtained from the red, green, blue, hue, and saturation components of the color images. The performance of the CBIR system was measured in terms of precision and recall. Initial results demonstrate the potential of the proposed methods with the best precision result of 70% obtained for the characterization of mixed tissue composition.

Pontine activation during focal status epilepticus secondary to hamartoma of the floor of the fourth ventricle

Epileptic seizures associated with hamartoma of the floor of the fourth ventricle (HFFV) are generally resistant to antiepileptic medication, may evolve into status epilepticus, and can respond favorably to surgical therapy. HFFV are rare, and during the neonatal or infantile period may be associated with repetitive and stereotyped attacks of hemifacial spasm, eye blinking, facial movements, head deviation and dysautonomic manifestations. Similarly, to gelastic seizures provoked by hypothalamic hamartomas, it has been suggested that these spells arise from within the HFFV, thus constituting a type of non-cortical seizure. We report an infant female patient that developed continuous left hemifacial attacks since she was 2-month-old, and that underwent presurgical investigation when she was 18-month-old. MRI disclosed a left sided HFFV, Video-EEG showed non-localizing and non-lateralizing findings, and SPECT aligned with MRI showed marked hyperperfusion within the hamartoma, spreading to ipsilateral cerebellar parenchyma and brainstem nuclei. Patient underwent lesionectomy and became seizure-free. We found two evidences on literature supporting the hypothesis of non-cortical seizures related to HFFV. The first, intra-cerebellar recordings surrounding hamartoma showed electrical activity related to seizures. The second, subtracted SPECT co-registered MRI showed hyperemia within hamartoma. The present report provides the third additional evidence. We found the involvement not only of the hamartoma, and pars of cerebellar hemisphere, but also an intense hyperemia involving brainstem nuclei during seizures. We believe that all these findings suggest a short subcortical network responsible for generating seizures in HFFV patients.

Using an image-extended relational database to support content-based image retrieval in a PACS

This paper presents a new Picture Archiving and Communication System (PACS), called cbPACS, which has content-based image retrieval capabilities. The cbPACS answers range and k-nearest- neighbor similarity queries, employing a relational database manager extended to support images. The images are compared through their features, which are extracted by an image-processing module and stored in the extended relational database. The database extensions were developed aiming at efficiently answering similarity queries by taking advantage of specialized indexing methods. The main concept supporting the extensions is the definition, inside the relational manager, of distance functions based on features extracted from the images. An extension to the SQL language enables the construction of an interpreter that intercepts the extended commands and translates them to standard SQL, allowing any relational database server to be used. By now, the system implemented works on features based on color distribution of the images through normalized histograms as well as metric histograms. Metric histograms are invariant regarding scale, translation and rotation of images and also to brightness transformations. The cbPACS is prepared to integrate new image features, based on texture and shape of the main objects in the image.

Effects of cannabidiol (CBD) on regional cerebral blood flow

Animal and human studies have suggested that cannabidiol (CBD) may possess anxiolytic properties, but how these effects are mediated centrally is unknown. The aim of the present study was to investigate this using functional neuroimaging. Regional cerebral blood flow (rCBF) was measured at rest using (99m)Tc-ECD SPECT in 10 healthy male volunteers, randomly divided into two groups of five subjects. Each subject was studied on two occasions, 1 week apart. In the first session, subjects were given an oral dose of CBD (400 mg) or placebo, in a double-blind procedure. SPECT images were acquired 90 min after drug ingestion. The Visual Analogue Mood Scale was applied to assess subjective states. In the second session, the same procedure was performed using the drug that had not been administered in the previous session. Within-subject between-condition rCBF comparisons were performed using statistical parametric mapping (SPM). CBD significantly decreased subjective anxiety and increased mental sedation, while placebo did not induce significant changes. Assessment of brain regions where anxiolytic effects of CBD were predicted a priori revealed two voxel clusters of significantly decreased ECD uptake in the CBD relative to the placebo condition (p<0.001, uncorrected for multiple comparisons). These included a medial temporal cluster encompassing the left amygdala-hippocampal complex, extending into the hypothalamus, and a second cluster in the left posterior cingulate gyrus. There was also a cluster of greater activity with CBD than placebo in the left parahippocampal gyrus (p<0.001). These results suggest that CBD has anxiolytic properties, and that these effects are mediated by an action on limbic and paralimbic brain areas.