Effects of emotional valence and intensity on cognitive and affective empathy after insula lesions.

The insula plays a central role in empathy. However, the complex structure of cognitive (CE) and affective empathy (AE) deficits following insular damage is not fully understood. In the present study, patients with insular lesions (n = 20) and demographically matched healthy controls (n = 24) viewed ecologically valid videos that varied in terms of valence and emotional intensity. The videos showed a person (target) narrating a personal life event. In CE conditions, subjects continuously rated the affective state of the target, while in AE conditions, they continuously rated their own affect. Mean squared error (MSE) assessed deviations between subject and target ratings. Patients differed from controls only in negative, low-intensity AE, rating their own affective state less negative than the target. This deficit was not related to trait empathy, neuropsychological or clinical parameters, or laterality of lesion. Empathic functions may be widely spared after insular damage in a naturalistic, dynamic setting, potentially due to the intact interpretation of social context by residual networks outside the lesion. The particular role of the insula in AE for negative states may evolve specifically in situations that bear higher uncertainty pointing to a threshold role of the insula in online ratings of AE.

Comparison of 68Ga-PSMA PET/CT and mp-MRI in regard to local staging for prostate cancer with histopathological results: A retrospective study.

BACKGROUND: Imaging modalities are used to diagnose and clinical grading of clinically significant prostate cancer. In this study, 68Ga-PSMA PET/CT (PSMA) and multiparametric prostate MRI (mp-MRI) were compared in regard to locating intraprostatic tumor and locoregional staging. METHODS: After ethics committee approval, a total of 49 patients with prostate cancer who had mp-MRI and PSMA before radical prostatectomy were included. Preoperative and postoperative PSA, transrectal ultrasound-guided prostate biopsy (TRUS-Bx) ISUP grade, radical prostatectomy ISUP grade, body mass index (BMI), TRUS prostate volume, mp-MRI tumor mapping, PSMAtumor mapping, pathologic tumor mapping, extraprostatic extension (EPE), seminal vesicle invasion (SVI), lymph node invasion (LNI), and bladder neck invasion (BNI)were retrospectively evaluated. Index tumor was located by uroradiologist, nuclear medicine specialist, and uropathologist on a 12-sector prostate pathology map and compared with each other in terms of accuracy and locoregional clinical staging. RESULTS: Mean age of the patients was 66.18 ± 6.67 years and the mean of preoperative PSA results was 21.11 ± 32.56 ng/ml. Nearly half of the patients' (44.9%) pathology was reported as ISUP grade 4 and 5% and 18.4% of patients were surgical margin positive. According to the pathological findings, 362 out of 588 sectors were tumor-positive, 174 out of 362 sectors were tumor-positive in mp-MRI, and 175 out of 362 sectors were tumor-positive in PSMA. Both PSMA and mp-MRI were comparable (p = 0.823) and accurate to detect the location of the intraprostatic index tumor (AUC = 0.66 vs. 0.69 respectively, p = 0.82). The sensitivity and the specificity of the PSMA and mp-MRI for localizing intraprostatic index tumors were 42.5% versus 49.5% and 90.7% versus 88.6% respectively. mp-MRI was more accurate than PSMA in terms of EPE (AUC = 0.8 vs. AUC = 0.57 respectively, p = 0.027) and both methods were comparable in terms of SVI (AUC = 0.75 vs. AUC = 0.75, p = 0.886) and BNI (AUC = 0.51 vs. AUC = 0.59, p = 0.597). PSMA and mp-MRI were comparable in terms of LNI (AUC = 0.76 vs. AUC = 0.64, p = 0.39). CONCLUSION: mp-MRI should be considered for its high accuracy in the diagnosis of EPE, especially before decision-making for nerve-sparing surgery in high-risk patients. Both imaging modalities were accurate for localizing intraprostatic index tumor. PSMA is accurate for detecting LNI.

Neuroanatomical and behavioural factors associated with the effectiveness of two weekly sessions of prism adaptation in the treatment of unilateral neglect.

Among the different interventions to alleviate the symptoms of unilateral neglect, prism adaptation (PA) appears especially promising. To elucidate the contribution of some neuroanatomical and behavioural factors to PA's effectiveness, we conducted a study combining neuropsychological and lesion mapping methods on a group of 19 neglect patients who underwent two sessions of PA during one week and assessed their improvement relative to the baseline until the following week (7-8 days later). Correlation analyses revealed a significant positive relationship between the magnitude of the proprioceptive after-effect and the improvement at the follow-up session in two perceptual tasks requiring motor responses. Conversely, no correlation was found between the proprioceptive after-effect and the improvement in a perceptual task with no motor involvement. This finding suggests that patients' potential to show a prism-related improvement in motor-related tasks might be indicated by the strength of their proprioceptive response (proprioceptive after-effect). As for the neuroanatomical basis of this relationship, subtraction analyses suggested that patients' improvement in perceptual tasks with high motor involvement might be facilitated by the integrity of temporo-parietal areas and the damage of frontal and subcortical areas.

Reprint of: Mapping human brain lesions and their functional consequences.

Neuroscience has a long history of inferring brain function by examining the relationship between brain injury and subsequent behavioral impairments. The primary advantage of this method over correlative methods is that it can tell us if a certain brain region is necessary for a given cognitive function. In addition, lesion-based analyses provide unique insights into clinical deficits. In the last decade, statistical voxel-based lesion behavior mapping (VLBM) emerged as a powerful method for understanding the architecture of the human brain. This review illustrates how VLBM improves our knowledge of functional brain architecture, as well as how it is inherently limited by its mass-univariate approach. A wide array of recently developed methods appear to supplement traditional VLBM. This paper provides an overview of these new methods, including the use of specialized imaging modalities, the combination of structural imaging with normative connectome data, as well as multivariate analyses of structural imaging data. We see these new methods as complementing rather than replacing traditional VLBM, providing synergistic tools to answer related questions. Finally, we discuss the potential for these methods to become established in cognitive neuroscience and in clinical applications.

The Dixon technique for MRI of the bone marrow.

Dixon sequences are established as a reliable MRI technique that can be used for problem-solving in the assessment of bone marrow lesions. Unlike other fat suppression methods, Dixon techniques rely on the difference in resonance frequency between fat and water and in a single acquisition, fat only, water only, in-phase and out-of-phase images are acquired. This gives Dixon techniques the unique ability to quantify the amount of fat within a bone lesion, allowing discrimination between marrow-infiltrating and non-marrow-infiltrating lesions such as focal nodular marrow hyperplasia. Dixon can be used with gradient echo and spin echo techniques, both two-dimensional and three-dimensional imaging. Another advantage is its rapid acquisition time, especially when using traditional two-point Dixon gradient echo sequences. Overall, Dixon is a robust fat suppression method that can also be used with intravenous contrast agents. After reviewing the available literature, we would like to advocate the implementation of additional Dixon sequences as a problem-solving tool during the assessment of bone marrow pathology.

Mapping human brain lesions and their functional consequences.

Neuroscience has a long history of inferring brain function by examining the relationship between brain injury and subsequent behavioral impairments. The primary advantage of this method over correlative methods is that it can tell us if a certain brain region is necessary for a given cognitive function. In addition, lesion-based analyses provide unique insights into clinical deficits. In the last decade, statistical voxel-based lesion behavior mapping (VLBM) emerged as a powerful method for understanding the architecture of the human brain. This review illustrates how VLBM improves our knowledge of functional brain architecture, as well as how it is inherently limited by its mass-univariate approach. A wide array of recently developed methods appear to supplement traditional VLBM. This paper provides an overview of these new methods, including the use of specialized imaging modalities, the combination of structural imaging with normative connectome data, as well as multivariate analyses of structural imaging data. We see these new methods as complementing rather than replacing traditional VLBM, providing synergistic tools to answer related questions. Finally, we discuss the potential for these methods to become established in cognitive neuroscience and in clinical applications.

'Whose atlas I use, his song I sing?' - The impact of anatomical atlases on fiber tract contributions to cognitive deficits after stroke.

Nowadays, different anatomical atlases exist for the anatomical interpretation of the results from neuroimaging and lesion analysis studies that investigate the contribution of white matter fiber tract integrity to cognitive (dys)function. A major problem with the use of different atlases in different studies, however, is that the anatomical interpretation of neuroimaging and lesion analysis results might vary as a function of the atlas used. This issue might be particularly prominent in studies that investigate the contribution of white matter fiber tract integrity to cognitive (dys)function. We used a single large-sample dataset of right brain damaged stroke patients with and without cognitive deficit (here: spatial neglect) to systematically compare the influence of three different, widely-used white matter fiber tract atlases (1 histology-based atlas and 2 DTI tractography-based atlases) on conclusions concerning the involvement of white matter fiber tracts in the pathogenesis of cognitive dysfunction. We both calculated the overlap between the statistical lesion analysis results and each long association fiber tract (topological analyses) and performed logistic regressions on the extent of fiber tract damage in each individual for each long association white matter fiber tract (hodological analyses). For the topological analyses, our results suggest that studies that use tractography-based atlases are more likely to conclude that white matter integrity is critical for a cognitive (dys)function than studies that use a histology-based atlas. The DTI tractography-based atlases classified approximately 10 times as many voxels of the statistical map as being located in a long association white matter fiber tract than the histology-based atlas. For hodological analyses on the other hand, we observed that the conclusions concerning the overall importance of long association fiber tract integrity to cognitive function do not necessarily depend on the white matter atlas used, but conclusions may vary as a function of atlas used at the level of individual fiber tracts. Moreover, these analyses revealed that hodological studies that express the individual extent of injury to each fiber tract as a binomial variable are more likely to conclude that white matter integrity is critical for a cognitive function than studies that express the individual extent of injury to each fiber tract as a continuous variable.

Impact of correction factors in human brain lesion-behavior inference.

Statistical voxel-based lesion-behavior mapping (VLBM) in neurological patients with brain lesions is frequently used to examine the relationship between structure and function of the healthy human brain. Only recently, two simulation studies noted reduced anatomical validity of this method, observing the results of VLBM to be systematically misplaced by about 16 mm. However, both simulation studies differed from VLBM analyses of real data in that they lacked the proper use of two correction factors: lesion size and "sufficient lesion affection." In simulation experiments on a sample of 274 real stroke patients, we found that the use of these two correction factors reduced misplacement markedly compared to uncorrected VLBM. Apparently, the misplacement is due to physiological effects of brain lesion anatomy. Voxel-wise topographies of collateral damage in the real data were generated and used to compute a metric for the inter-voxel relation of brain damage. "Anatomical bias" vectors that were solely calculated from these inter-voxel relations in the patients' real anatomical data, successfully predicted the VLBM misplacement. The latter has the potential to help in the development of new VLBM methods that provide even higher anatomical validity than currently available by the proper use of correction factors. Hum Brain Mapp 38:1692-1701, 2017. © 2017 Wiley Periodicals, Inc.

The impact of lesion location on dysphagia incidence, pattern and complications in acute stroke. Part 2: Oropharyngeal residue, swallow and cough response, and pneumonia.

BACKGROUND AND PURPOSE: Dysphagia is a well-known complication of acute stroke. Given the complexity of cerebral swallowing control it is still difficult to predict which patients are likely to develop swallowing dysfunction based on their neuroimaging. In Part 2 of a comprehensive voxel-based imaging study, whether the location of a stroke lesion can be correlated with further dysfunctional swallowing patterns, pulmonary protective reflexes and pneumonia was evaluated. METHODS: In all, 200 acute stroke cases were investigated applying flexible endoscopic evaluation of swallowing within 96 h from admission. Lesions were mapped using patients' computed tomography/magnetic resonance images and these were registered to a standard space. The percentage of lesioned volume of 137 anatomically defined brain regions was determined on a voxel basis (FSL5.0). Region-specific odds ratios (ORs) were calculated with respect to the presence of oropharyngeal residue, delayed swallow response, insufficient cough reflex and occurrence of pneumonia during hospital stay. Colour-coded lesion location maps of brain regions with significant ORs were created (P < 0.05). RESULTS: Lesion maps for residue and impaired swallow response depicted parietal-temporal areas of the right hemisphere. Limbic structures in the right hemisphere and sensory regions on the left were associated with cough reflex disturbance. There was no overlap of lesion maps for impaired swallow response and insufficient cough reflex or pneumonia, but substantial overlap between the last two conditions. CONCLUSIONS: This study gives new insights on the cortical representation of single components of swallowing and airway protection behaviours. The lesion model may help to risk-stratify patients for dysphagia and pneumonia based on their brain scan.

Testing the dual-pathway model for auditory processing in human cortex.

Analogous to the visual system, auditory information has been proposed to be processed in two largely segregated streams: an anteroventral ("what") pathway mainly subserving sound identification and a posterodorsal ("where") stream mainly subserving sound localization. Despite the popularity of this assumption, the degree of separation of spatial and non-spatial auditory information processing in cortex is still under discussion. In the present study, a statistical approach was implemented to investigate potential behavioral dissociations for spatial and non-spatial auditory processing in stroke patients, and voxel-wise lesion analyses were used to uncover their neural correlates. The results generally provided support for anatomically and functionally segregated auditory networks. However, some degree of anatomo-functional overlap between "what" and "where" aspects of processing was found in the superior pars opercularis of right inferior frontal gyrus (Brodmann area 44), suggesting the potential existence of a shared target area of both auditory streams in this region. Moreover, beyond the typically defined posterodorsal stream (i.e., posterior superior temporal gyrus, inferior parietal lobule, and superior frontal sulcus), occipital lesions were found to be associated with sound localization deficits. These results, indicating anatomically and functionally complex cortical networks for spatial and non-spatial auditory processing, are roughly consistent with the dual-pathway model of auditory processing in its original form, but argue for the need to refine and extend this widely accepted hypothesis.

Paired comparison of the sensitivity and specificity of multispectral digital skin lesion analysis and reflectance confocal microscopy in the detection of melanoma in vivo: A cross-sectional study.

BACKGROUND: Several technologies have been developed to aid dermatologists in the detection of melanoma in vivo including dermoscopy, multispectral digital skin lesion analysis (MDSLA), and reflectance confocal microscopy (RCM). To our knowledge, there have been no studies directly comparing MDSLA and RCM. OBJECTIVE: We conducted a repeated measures analysis comparing the sensitivity and specificity of MDSLA and RCM in the detection of melanoma (n = 55 lesions from 36 patients). METHODS: Study patients (n = 36) with atypical-appearing pigmented lesions (n = 55) underwent imaging by both RCM and MDSLA. Lesions were biopsied and analyzed by histopathology. RESULTS: RCM exhibited superior test metrics (P = .001, McNemar test) compared with MDSLA. Respectively, sensitivity measures were 85.7% and 71.4%, and specificity rates were 66.7% and 25.0%. LIMITATIONS: The sample size was relatively small and was collected from only one dermatologist's patient base; there was some degree of dermatopathologist interobserver variability; and only one confocalist performed the RCM image evaluations. CONCLUSION: RCM is a useful adjunct during clinical assessment of in vivo lesions suspicious for melanoma or those requiring re-excision because of high level of dysplasia or having features consistent with an atypical melanocytic nevus with severe cytologic atypia.

Reliance on functional resting-state network for stable task control predicts behavioral tendency for cooperation.

Humans display individual variability in cooperative behavior. While an ever-growing body of research has investigated the neural correlates of task-specific cooperation, the mechanisms by which situation-independent, stable differences in cooperation render behavior consistent across a wide range of situations remain elusive. Addressing this issue, we show that the individual tendency to behave in a prosocial or individualistic manner can be predicted from the functional resting-state connectome. More specifically, connections of the cinguloopercular network which supports goal-directed behavior encode cooperative tendency. Effects of virtual lesions to this network on the efficacy of information exchange throughout the brain corroborate our findings. These results shed light on the neural mechanisms underlying individualists' and prosocials' habitual social decisions by showing that reliance on the cinguloopercular task-control network predicts stable cooperative behavior. Based on this evidence, we provide a unifying framework for the interpretation of functional imaging and behavioral studies of cooperative behavior.

The impact of lesion location on dysphagia incidence, pattern and complications in acute stroke. Part 1: dysphagia incidence, severity and aspiration.

BACKGROUND AND PURPOSE: Although early identification of patients at risk for dysphagia is crucial in acute stroke care, predicting whether a particular patient is likely to have swallowing problems based on the brain scan is difficult because a comprehensive model of swallowing control is missing. In this study whether stroke location is associated with dysphagia incidence, severity and the occurrence of penetration or aspiration was systematically evaluated relying on a voxel-based imaging analysis approach. METHODS: Two hundred acute stroke patients were investigated applying fiberoptic endoscopic evaluation of swallowing within 96 h from admission. Lesion masks were obtained from each patient's brain scan and registered to standard space. The percentage of lesioned volume of 137 atlas-based brain regions was determined in each case. Region-specific odds ratios were afterwards calculated with respect to presence of dysphagia, its severity and occurrence of penetration or aspiration. RESULTS: In all, 165 patients were diagnosed with dysphagia, 80 of whom had severe swallow impairment. For each investigated item there were significant differences of regional percentage infarction in distinct brain areas between affected patients and those who did not present with that specific dysfunction. In particular, right hemispheric lesions of the pre- and post-central gyri, opercular region, supramarginal gyrus and respective subcortical white matter tracts were related to dysphagia, with post-central lesions being especially associated with severe swallowing impairment. CONCLUSIONS: Distinct brain lesion locations are related to the incidence, severity and pattern of swallowing dysfunction.

Mental paper folding performance following penetrating traumatic brain injury in combat veterans: a lesion mapping study.

Mental paper folding is a complex measure of visuospatial ability involving a coordinated sequence of mental transformations and is often considered a measure of mental ability. The literature is inconclusive regarding the precise neural architecture that underlies performance. We combined the administration of the Armed Forces Qualification Test boxes subtest measuring mental paper folding ability, with a voxel-based lesion symptom mapping approach to identify brain regions associated with impaired mental paper folding ability. Using a large sample of subjects with penetrating traumatic brain injury and defined lesions studied over 2 time points, roughly 15 and 35 years post-injury, enabled us to answer the causal questions regarding mental paper folding impairment. Our results revealed that brain injury significantly exacerbates the decline of performance on mental paper folding tasks over time. Our study adds novel neuropsychological and neuroimaging support for parietal lobe involvement; specifically the right inferior parietal lobule (Broadmann's Area [BA] 40) and the left parahippocampal region (BAs 19, 36). Both areas were consistently associated with mental paper folding performance and demonstrate that the right parietal lobe and the left parahippocampal gyrus play an integral role in mental paper folding tasks.

Automatic lesion detection at Multiple Sclerosis patients - Comparison of 2D- and 3D-FLAIR-datasets.

BACKGROUND: Multiple Sclerosis (MS) is a common autoimmune inflammatory disease of the central nervous system (CNS). Magnetic Resonance Imaging (MRI) allows a sensitive assessment of the CNS and is established for diagnostic, prognostic and (therapy-) monitoring purposes. Especially lesion counting in T2- or Fluid Attenuated Inversion Recovery (FLAIR)-weighted images plays a decisive role in clinical routine. Software-packages allowing an automatic evaluation of image data are increasingly established aiming a faster and improved workflow. These programs allow e.g. the counting, spatial attribution and volumetry of MS-lesions in FLAIR-weighted images. Research has shown that 3D-FLAIR-sequences are superior to 2D-FLAIR-sequences in visual evaluation of lesion burden in MS. An influence on the automatic analysis is expectable but not yet systematically studied. This work will therefore investigate the influence of 2D- and 3D datasets on the results of an automatic assessment. MATERIAL AND METHODS: In this prospective study, 80 Multiple Sclerosis patients underwent a clinically indicated routine MRI examination. The clinical routine protocol already including a 3D-FLAIR sequence was adapted by an additional 2D-FLAIR sequence also conform to the 2021 MAGNIMS-CMSCNAIMS consensus recommendations. To obtain a quantitative analysis for assessment of amount, dissemination and volume of the lesions, the acquired MR images were post-processed using the CE-certified Software mdbrain (mediaire, Berlin, Germany). The resulting data were statistically analysed using the paired t-test for normally distributed data and the Wilcoxon-signed-rank-test for not normally distributed data respectively. Demographic data and data such as the subtype, duration, severity and therapy of the disease were collected, pseudonymized and evaluated. RESULTS: There is a significant difference concerning the total number and lesion volume with more lesions being detected (2D: 29.7, +/- 20.22 sd; 3D: 40.1 +/- 31.67 sd; p < 0.0001) but lower total volume (2D: 6.24 +/- 6.11 sd; 3D: 5.39 +/- 6.37 sd; p < 0.0001) when using the 3D- sequence. Especially significantly more small lesions in the unspecific white matter and infratentorial region were detected by using the 3D-FLAIR sequence (p < 0.0001) compared to the 2D-FLAIR image. Main reason for the lower total volume in the 3D-FLAIR sequence was the calculated volume for periventricular lesions which was significantly beneath the calculated volume from the 2D-FLAIR sequence (p < 0.0001). CONCLUSION: Automatic lesion counting and volumetry is feasible with both 2D- and 3D-weightend FLAIR images. Still, it leads to partly significant differences even between two sequences that both are conform to the 2021 MAGNIMS-CMSCNAIMS consensus recommendations. This study contributes valuable insights into the impact of using different input data from the same patient for automated MS lesion evaluation.

ESFPNet: Efficient Stage-Wise Feature Pyramid on Mix Transformer for Deep Learning-Based Cancer Analysis in Endoscopic Video.

For patients at risk of developing either lung cancer or colorectal cancer, the identification of suspect lesions in endoscopic video is an important procedure. The physician performs an endoscopic exam by navigating an endoscope through the organ of interest, be it the lungs or intestinal tract, and performs a visual inspection of the endoscopic video stream to identify lesions. Unfortunately, this entails a tedious, error-prone search over a lengthy video sequence. We propose a deep learning architecture that enables the real-time detection and segmentation of lesion regions from endoscopic video, with our experiments focused on autofluorescence bronchoscopy (AFB) for the lungs and colonoscopy for the intestinal tract. Our architecture, dubbed ESFPNet, draws on a pretrained Mix Transformer (MiT) encoder and a decoder structure that incorporates a new Efficient Stage-Wise Feature Pyramid (ESFP) to promote accurate lesion segmentation. In comparison to existing deep learning models, the ESFPNet model gave superior lesion segmentation performance for an AFB dataset. It also produced superior segmentation results for three widely used public colonoscopy databases and nearly the best results for two other public colonoscopy databases. In addition, the lightweight ESFPNet architecture requires fewer model parameters and less computation than other competing models, enabling the real-time analysis of input video frames. Overall, these studies point to the combined superior analysis performance and architectural efficiency of the ESFPNet for endoscopic video analysis. Lastly, additional experiments with the public colonoscopy databases demonstrate the learning ability and generalizability of ESFPNet, implying that the model could be effective for region segmentation in other domains.

Lesion Analysis in PERCIST 1.0: Clinical Ease versus Research Requisite-Where Does the Balance Exist?

Background Semiqualitative parameter SUVmax has been the most frequently used semiquantitative positron emission tomography (PET) parameter for response evaluation, but only metabolic activity of a single (most metabolic) lesion is predicted. Newer response parameters such as tumor lesion glycolysis (TLG) incorporating lesions' metabolic volume or whole-body metabolic tumor burden (MTBwb) are being explored for response evaluation. Evaluation and comparison of response with different semiquantitative PET parameters such as SUVmax and TLG in most metabolic lesion, multiple lesions (max of five), and MTBwb in advanced non-small cell lung cancer (NSCLC) patients were made. The different PET parameters were analyzed for response evaluation, overall survival (OS), and progression-free survival (PFS). Methods 18 F-FDG-PET/CT (18-fluorine-fluorodeoxyglucose positron emission tomography/computed tomography) imaging was performed in 23 patients (M = 14, F = 9, mean age = 57.6 years) with stage IIIB-IV advanced NSCLC before initiation of therapy with oral estimated glomerular filtration rate-tyrosine kinase inhibitor for early and late response evaluation. The quantitative PET parameters such as SUVmax and TLG were measured in single (most metabolic) lesion, multiple lesions, and MTBwb. The parameters SUVmax, TLG, and MTBwb were compared for early and late response evaluation and analyzed for OS and PFS Results No significant difference in change in response evaluation was seen in patients evaluated with most metabolic lesion, multiple lesions, or MTBwb. Difference in early (DC 22, NDC 1) and late (DC 20, NDC 3) response evaluation was seen that remained unchanged when lesions were measured in terms of number of lesions or the MTBwb. The early imaging was seen to be statistically significant to the OS compared with late imaging. Conclusions Single (most metabolic) lesion shows similar disease response and OS to multiple lesions and MTBwb. Response evaluation by late imaging offered no significant advantage compared with early imaging. Thus, early response evaluation with SUVmax parameter offers a good balance between clinical ease and research requisition.

Multiclass skin lesion localization and classification using deep learning based features fusion and selection framework for smart healthcare.

BACKGROUND: The idea of smart healthcare has gradually gained attention as a result of the information technology industry's rapid development. Smart healthcare uses next-generation technologies i.e., artificial intelligence (AI) and Internet of Things (IoT), to intelligently transform current medical methods to make them more efficient, dependable and individualized. One of the most prominent uses of telemedicine and e-health in medical image analysis is teledermatology. Telecommunications technologies are used in this industry to send medical information to professionals. Teledermatology is a useful method for the identification of skin lesions, particularly in rural locations, because the skin is visually perceptible. One of the most recent tools for diagnosing skin cancer is dermoscopy. To classify skin malignancies, numerous computational approaches have been proposed in the literature. However, difficulties still exist i.e., lesions with low contrast, imbalanced datasets, high level of memory complexity, and the extraction of redundant features. METHODS: In this work, a unified CAD model is proposed based on a deep learning framework for skin lesion segmentation and classification. In the proposed approach, the source dermoscopic images are initially pre-processed using a contrast enhancement based modified bio-inspired multiple exposure fusion approach. In the second stage, a custom 26-layered convolutional neural network (CNN) architecture is designed to segment the skin lesion regions. In the third stage, four pre-trained CNN models (Xception, ResNet-50, ResNet-101 and VGG16) are modified and trained using transfer learning on the segmented lesion images. In the fourth stage, the deep features vectors are extracted from all the CNN models and fused using the convolutional sparse image decomposition fusion approach. In the fifth stage, the univariate measurement and Poisson distribution feature selection approach is used for the best features selection for classification. Finally, the selected features are fed to the multi-class support vector machine (MC-SVM) for the final classification. RESULTS: The proposed approach employed to the HAM10000, ISIC2018, ISIC2019, and PH2 datasets and achieved an accuracy of 98.57%, 98.62%, 93.47%, and 98.98% respectively which are better than previous works. CONCLUSION: When compared to renowned state-of-the-art methods, experimental results show that the proposed skin lesion detection and classification approach achieved higher performance in terms of both visually and enhanced quantitative evaluation with enhanced accuracy.

Lateralized deficits in arousal processing after insula lesions: Behavioral and autonomic evidence.

A large body of evidence ascribes a pivotal role in emotion processing to the insular cortex. However, the complex structure and lateralization of emotional deficits following insular damage are not understood. Here, we investigated emotional ratings of valence and arousal and skin conductance responses (SCR) to a graded series of emotionally arousing scenes in patients with left (n = 10) or right (n = 9) insular damage and in healthy controls (n = 18). We found a significant reduction in overall SCRs, arousal ratings and valence extremity scores in right-lesioned patients, as compared to left-lesioned patients and healthy controls. The degree of right insular damage was significantly correlated with the degree of arousal, SCR and extremity attenuation. Additional analyses of correlations between subjective arousal ratings resp. SCR and normative arousal ratings revealed that both lesion groups had evaluative and physiological difficulties to discover changes in stimulus arousal. Although no group differences emerged on overall ratings of valence, analysis of correlations between subjective and normative valence ratings displayed markedly reduced accuracy in right-lesioned patients, as compared to left-lesioned patients and healthy controls. Our findings support the hypothesis that the left and right insulae subserve different functions in emotion processing, potentially due to asymmetrical representations of autonomic information in the left and right human forebrain. The right insula may serve as integral node for sympathetic arousal and cognitive-affective processing.

Knowledge distillation approach towards melanoma detection.

Melanoma is regarded as the most threatening among all skin cancers. There is a pressing need to build systems which can aid in the early detection of melanoma and enable timely treatment to patients. Recent methods are geared towards machine learning based systems where the task is posed as image recognition, tag dermoscopic images of skin lesions as melanoma or non-melanoma. Even though these methods show promising results in terms of accuracy, they are computationally quite expensive to train, that questions the ability of these models to be deployable in a clinical setting or memory constraint devices. To address this issue, we focus on building simple and performant models having few layers, less than ten compared to hundreds. As well as with fewer learnable parameters, 0.26 million (M) compared to 42.5 M using knowledge distillation with the goal to detect melanoma from dermoscopic images. First, we train a teacher model using a ResNet-50 to detect melanoma. Using the teacher model, we train the student model known as Distilled Student Network (DSNet) which has around 0.26 M parameters using knowledge distillation achieving an accuracy of 91.7%. We compare against ImageNet pre-trained models such MobileNet, VGG-16, Inception-V3, EfficientNet-B0, ResNet-50 and ResNet-101. We find that our approach works well in terms of inference runtime compared to other pre-trained models, 2.57 s compared to 14.55 s. We find that DSNet (0.26 M parameters), which is 15 times smaller, consistently performs better than EfficientNet-B0 (4 M parameters) in both melanoma and non-melanoma detection across Precision, Recall and F1 scores.