Computed Tomography Classification Research Articles

Resting-State EEG Signature of Early Consciousness Recovery in Comatose Patients with Traumatic Brain Injury.

Resting-state electroencephalography (rsEEG) is usually obtained to assess seizures in comatose patients with traumatic brain injury (TBI). We aim to investigate rsEEG measures and their prediction of early recovery of consciousness in patients with TBI. This is a retrospective study of comatose patients with TBI who were admitted to a trauma center (October 2013 to January 2022). Demographics, basic clinical data, imaging characteristics, and EEGs were collected. We calculated the following using 10-min rsEEGs: power spectral density, permutation entropy (complexity measure), weighted symbolic mutual information (wSMI, global information sharing measure), Kolmogorov complexity (Kolcom, complexity measure), and heart-evoked potentials (the averaged EEG signal relative to the corresponding QRS complex on electrocardiography). We evaluated the prediction of consciousness recovery before hospital discharge using clinical, imaging, and rsEEG data via a support vector machine. We studied 113 of 134 (84%) patients with rsEEGs. A total of 73 (65%) patients recovered consciousness before discharge. Patients who recovered consciousness were younger (40 vs. 50years, p = 0.01). Patients who recovered also had higher Kolcom (U = 1688, p = 0.01), increased beta power (U = 1,652 p = 0.003) with higher variability across channels (U = 1534, p = 0.034) and epochs (U = 1711, p = 0.004), lower delta power (U = 981, p = 0.04), and higher connectivity across time and channels as measured by wSMI in the theta band (U = 1636, p = 0.026; U = 1639, p = 0.024) than those who did not recover. The area under the receiver operating characteristic curve for rsEEG was higher than that for clinical data (using age, motor response, pupil reactivity) and higher than that for the Marshall computed tomography classification (0.69 vs. 0.66 vs. 0.56, respectively; p < 0.001). We describe the rsEEG signature in recovery of consciousness prior to discharge in comatose patients with TBI. rsEEG measures performed modestly better than the clinical and imaging data in predicting recovery.

Utility of serum chemokine-like factor 1 as a biomarker of severity and prognosis after severe traumatic brain injury: A prospective observational study.

Chemokine-like factor 1 (CKLF1) may be involved in the inflammatory response and secondary brain injury after severe traumatic brain injury (sTBI). We determined serum CKLF1 levels of sTBI patients to further investigate the correlation of CKLF1 levels with disease severity, functional prognosis, and 180-day mortality of sTBI. Serum CKLF1 levels were measured at admission in 119 sTBI patients and at entry into study in 119 healthy controls. Serum CKLF levels of 50 patients were also quantified at days 1-3, 5, and 7 after admission. Glasgow coma scale (GCS) scores and Rotterdam computerized tomography (CT) classification were utilized to assess disease severity. Extended Glasgow outcome scale (GOSE) scores were recorded to evaluate function prognosis at 180 days after sTBI. Relations of serum CKLF1 levels to 180-day poor prognosis (GOSE scores of 1-4) and 180-day mortality were analyzed using univariate analysis, followed by multivariate analysis. Receiver-operating characteristic (ROC) curve was built to investigate prognostic predictive capability. Serum CKLF1 levels of sTBI patients increased at admission, peaked at day 2, and then gradually decreased; they were significantly higher during the 7 days after sTBI than in healthy controls. Differences of areas under ROC curve (areas under the curve [AUCs]) were not significant among the six time points. Multivariate analysis showed that serum CKLF1 levels were independently correlated with GCS scores, Rotterdam CT classification, and GOSE scores. Serum CKLF1 levels were significantly higher in non-survivors than in survivors and in poor prognosis patients than in good prognosis patients. Serum CKLF1 levels independently predicted 180-day poor prognosis and 180-day mortality, and had high 180-day prognosis and mortality predictive abilities, and their AUCs were similar to those of GCS scores and Rotterdam CT classification. Combination model containing serum CKLF1, GCS scores, and Rotterdam CT classification performed more efficiently than any of them alone in predicting mortality and poor prognosis. The models were visually described using nomograms, which were comparatively stable under calibration curve and were relatively of clinical benefit under decision curve. Serum CKLF1 levels are significantly associated with disease severity, poor 180-day prognosis, and 180-day mortality in sTBI patients. Hence, complement CKLF1 may serve as a potential prognostic biomarker of sTBI.

Compatible with fibrotic hypersensitivity pneumonitis on high-resolution computed tomography: from the ATS/JRS/ALAT 2020 hypersensitivity pneumonitis guidelines.

In compatible with fibrotic hypersensitivity pneumonitis (HP) of the computed tomography (CT) classification using the American Thoracic Society (ATS)/Japanese Respiratory Society (JRS)/Latin American Thoracic Association (ALAT) HP guidelines, the lung fibrosis pattern was classified as either a usual interstitial pneumonia (UIP) pattern or a diffuse ground-glass opacity (GGO) pattern with subtle fibrosis. We investigated whether patients with the same imaging classification had different disease progression. We also attempted to reclassify these patients using the CHEST HP guidelines. Patients with fibrotic HP who had compatible CT pattern in the ATS/JRS/ALAT classification were investigated retrospectively. With 62 patients in the UIP pattern group and 25 patients in the diffuse GGO pattern group, 87 patients with fibrotic HP had compatible pattern on CT. Annual forced vital capacity changes in the UIP pattern group and diffuse GGO pattern group were -2.7% and +3.3% (P=0.004), respectively. The 5-year survival rates in the UIP pattern group and diffuse GGO pattern group were 86% and 100% (P=0.02). In UIP pattern group in the ATS/JRS/ALAT classification, 27% patients were classified as typical fibrotic HP pattern in the CHEST guidelines. In the diffuse GGO pattern group, 52% patients were classified as typical pattern of fibrotic HP. In the CHEST guidelines, more patients in the GGO pattern were classified as typical pattern compared with those in the UIP pattern (P=0.02). The two patterns in compatible with fibrotic HP of CT classification using the ATS/JRS/ALAT HP guidelines had different disease progression. Typical patterns were more frequent in the CHEST guidelines than the ATS/JRS/ALAT guidelines.

Imaging Findings in Acute Traumatic Brain Injury: a National Institute of Neurological Disorders and Stroke Common Data Element-Based Pictorial Review and Analysis of Over 4000 Admission Brain Computed Tomography Scans from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study.

In 2010, the National Institute of Neurological Disorders and Stroke (NINDS) created a set of common data elements (CDEs) to help standardize the assessment and reporting of imaging findings in traumatic brain injury (TBI). However, as opposed to other standardized radiology reporting systems, a visual overview and data to support the proposed standardized lexicon are lacking. We used over 4000 admission computed tomography (CT) scans of patients with TBI from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) study to develop an extensive pictorial overview of the NINDS TBI CDEs, with visual examples and background information on individual pathoanatomical lesion types, up to the level of supplemental and emerging information (e.g., location and estimated volumes). We documented the frequency of lesion occurrence, aiming to quantify the relative importance of different CDEs for characterizing TBI, and performed a critical appraisal of our experience with the intent to inform updating of the CDEs. In addition, we investigated the co-occurrence and clustering of lesion types and the distribution of six CT classification systems. The median age of the 4087 patients in our dataset was 50 years (interquartile range, 29-66; range, 0-96), including 238 patients under 18 years old (5.8%). Traumatic subarachnoid hemorrhage (45.3%), skull fractures (37.4%), contusions (31.3%), and acute subdural hematoma (28.9%) were the most frequently occurring CT findings in acute TBI. The ranking of these lesions was the same in patients with mild TBI (baseline Glasgow Coma Scale [GCS] score 13-15) compared with those with moderate-severe TBI (baseline GCS score 3-12), but the frequency of occurrence was up to three times higher in moderate-severe TBI. In most TBI patients with CT abnormalities, there was co-occurrence and clustering of different lesion types, with significant differences between mild and moderate-severe TBI patients. More specifically, lesion patterns were more complex in moderate-severe TBI patients, with more co-existing lesions and more frequent signs of mass effect. These patients also had higher and more heterogeneous CT score distributions, associated with worse predicted outcomes. The critical appraisal of the NINDS CDEs was highly positive, but revealed that full assessment can be time consuming, that some CDEs had very low frequencies, and identified a few redundancies and ambiguity in some definitions. Whilst primarily developed for research, implementation of CDE templates for use in clinical practice is advocated, but this will require development of an abbreviated version. In conclusion, with this study, we provide an educational resource for clinicians and researchers to help assess, characterize, and report the vast and complex spectrum of imaging findings in patients with TBI. Our data provides a comprehensive overview of the contemporary landscape of TBI imaging pathology in Europe, and the findings can serve as empirical evidence for updating the current NINDS radiologic CDEs to version 3.0.

Prediction for the prognosis of diffuse axonal injury using automated pupillometry

ObjectivePrevious studies have reported various predictive indicators of diffuse axonal injury (DAI), but no consensus has not been reached. Although the efficiency of automated pupillometry in patients with consciousness disorder has been widely reported, there are few reports of its use in patients with DAI. This study aimed to investigate the significance of pupillary findings in predicting the prognosis of DAI. Patients and MethodsWe included patients admitted to our center with a diagnosis of DAI from June 1, 2021 to June 30, 2022. Pupillary findings in both eyes were quantitatively measured by automated pupillometry every 2 hours after admission. We statistically examined the correlations between automated pupillometry parameters, the patients’ characteristics, and outcomes such as the Glasgow Outcome Scale Extended (GOSE) after 6 months from injury, the time to follow command, and so on. ResultsAmong 22 patients included in this study, five had oculomotor nerve palsy. Oculomotor nerve palsy was correlated with all outcomes, whereas Marshall computed tomography (CT) classification, Injury severity score (ISS) and DAI grade were correlated with few outcomes. Some of the automated pupillometry parameters were significantly correlated with GOSE at 6 months after injury, and many during the first 24 hours of measurement were correlated with the time to follow command. Most of these results were not affected by adjustment using sedation period, ISS or Marshall CT classification. A subgroup analysis of patients without oculomotor nerve palsy revealed that many of the automated pupillometry parameters during the first 24 hours of measurement were significantly correlated with most of the outcomes. The cutoff values that differentiated a good prognosis (GOSE 5–8) from a poor prognosis (GOSE 1–4) were constriction velocity (CV) 1.43 (AUC = 0.81(0.62–1), p = 0.037) and maximum constriction velocity (MCV) 2.345 (AUC = 0.78 (0.58–0.98), p = 0.04). The cutoff values that differentiated the time to follow command into within 7 days and over 8 days were percentage of constriction 8 (AUC = 0.89 (0.68–1), p = 0.011), CV 0.63 (AUC = 0.92 (0.78–1), p = 0.013), MCV 0.855 (AUC = 0.9 (0.74–1), p = 0.017) and average dilation velocity 0.175 (AUC = 0.95 (0.86–1), p = 0.018). ConclusionsThe present results indicate that pupillary findings in DAI are a strong predictive indicator of the prognosis, and that quantitative measurement of them using automated pupillometry could facilitate enhanced prediction for the prognosis of DAI.

CT scan pancreatic cancer segmentation and classification using deep learning and the tunicate swarm algorithm.

Pancreatic cancer (PC) is a very lethal disease with a low survival rate, making timely and accurate diagnoses critical for successful treatment. PC classification in computed tomography (CT) scans is a vital task that aims to accurately discriminate between tumorous and non-tumorous pancreatic tissues. CT images provide detailed cross-sectional images of the pancreas, which allows oncologists and radiologists to analyse the characteristics and morphology of the tissue. Machine learning (ML) approaches, together with deep learning (DL) algorithms, are commonly explored to improve and automate the performance of PC classification in CT scans. DL algorithms, particularly convolutional neural networks (CNNs), are broadly utilized for medical image analysis tasks, involving segmentation and classification. This study explores the design of a tunicate swarm algorithm with deep learning-based pancreatic cancer segmentation and classification (TSADL-PCSC) technique on CT scans. The purpose of the TSADL-PCSC technique is to design an effectual and accurate model to improve the diagnostic performance of PC. To accomplish this, the TSADL-PCSC technique employs a W-Net segmentation approach to define the affected region on the CT scans. In addition, the TSADL-PCSC technique utilizes the GhostNet feature extractor to create a group of feature vectors. For PC classification, the deep echo state network (DESN) model is applied in this study. Finally, the hyperparameter tuning of the DESN approach occurs utilizing the TSA which assists in attaining improved classification performance. The experimental outcome of the TSADL-PCSC method was tested on a benchmark CT scan database. The obtained outcomes highlighted the significance of the TSADL-PCSC technique over other approaches to PC classification.

A rare presentation of bilateral emphysematous pyelonephritis with concurrent lung abscess: A case report

Bilateral emphysematous pyelonephritis (EPN) is a rare complication in uncontrolled diabetes mellitus patients and linked to a high patient mortality rate [1]. We present a case of a 62-year-old female patient with bilateral EPN who also developed a concurrent lung abscess. She came with high-grade fever, lethargy, and left flank pain. She had left lumbar tenderness and right-sided chest crepitations. Upon hospitalization, the patient developed septic shock. However, severity of her clinical presentation does not correlate with chest X-ray of right sided pneumonic changes and cavitation of the lungs on admission. Abdominopelvic computed tomography (CT) imaging revealed extensive left EPN (Huang-Tseng CT classification: class 4), early right EPN (Huang-Tseng CT classification: class 2), and right middle lobe lung abscess, which was treated with intravenous antibiotics, drainage, and optimal glycaemic management. This was followed by the resolution of sepsis and improvement in clinical status.

Observer- and radiomics model-based computed tomography classification of suppurative versus tuberculous lymphadenitis complicated with nodal necrosis of the neck in children.

Computed tomography (CT) can be used for the early detection of lymphadenitis. Radiomics is able to identify a large amount of hidden information from images. However, few CT-based radiomics studies on cervical lymphadenitis in children have been published. This study aimed to investigate the role of visual CT analysis and CT radiomics in differentiating cervical suppurative node necrosis from tuberculous node necrosis in pediatric patients. A total of 101 patients with cervical suppurative lymphadenitis (n=52) or cervical tuberculous lymphadenitis (n=49) were included. Clinical data and CT images were retrieved for analysis. For visual observation, 11 major CT features were identified for univariate and multivariate analyses. For radiomics analysis, image segmentation, feature value extraction, and dimension reduction, feature selection and the construction of radiomics-based models were performed through the RadCloud platform. For the visual observation, significant differences were found between the two groups, including the short diameter of the largest necrotic lymph node (P=0.03), sharp border of the node (P=0.02), fusion of nodes (P=0.02), regular silhouette of the necrotic area (P=0.001), multilocular necrotic area (P=0.02), node calcification (P=0.004), and enhancement degree of the nodal nonnecrotic area (P=0.01). No feature was found to be an independent predictor for suppurative or tuberculous lymphadenitis (P>0.05 for all features). Concerning the radiomics analysis, after feature value extraction and dimension reduction, nine related features were selected. The support vector machine classifier achieved high diagnostic performance in distinguishing suppurative from tuberculous lymphadenitis. The area under the curve, accuracy, sensitivity, and specificity of the support vector machine model test set were 0.89 (95% confidence interval: 0.72-1.00), 0.88, 0.78, and 0.90, respectively. Compared to observer-based CT image analyses, radiomics model-based CT image analyses exhibit better performance in the differential diagnosis of cervical suppurative and tuberculous lymphadenitis complicated with nodal necrosis in children.

Utility of 3D Planning Software in Understanding Residual Proximal Femoral Deformity for Planning of Revision Hip Arthroscopy

Background: During the early evolution of femoroacetabular impingement (FAI) treatment, undercorrection of femoral deformity was a leading cause of hip arthroscopy failures. As the pendulum has swung, overresection of femoral deformity has increased in prevalence as a cause of persistent hip pain after arthroscopy. Computed tomography (CT) scans are increasingly being used in hip arthroscopy for preoperative planning purposes and may allow for improved 3-dimensional (3D) assessment of complex femoral deformities after previous femoroplasty. Purpose: To assess whether CT scans provide additional utility over standard radiographs in understanding proximal femoral morphology in patients being evaluated for revision hip arthroscopy after previous femoroplasty in the setting of FAI. Study Design: Cohort study (Diagnosis); Level of evidence, 3. Methods: Preoperative CT scans and standard radiographs were obtained in 80 patients who underwent revision hip arthroscopy for FAI. The anteroposterior and Dunn radiographic views were used to assess patients for residual proximal femoral deformity and were compared with the CT scan views using a commercially available software program. Determinations of underresection were made using alpha angle, while overresection was determined according to a previously described technique. Chi-square tests were performed to determine statistical significance between radiographic and CT classifications of overresection, underresection, and concomitant over- and underresection. A kappa value was calculated to determine the agreement between measurements on the radiographs and CT scans. Results: There were 30 patients (37.5%) for whom the CT scans revealed information about femoral morphology that was not detected on the radiographs. The kappa value of agreement was 0.28 between CT and radiographic measurements. Underresected cams were detected in 30 patients (37.5%) on CT scans versus 17 patients (21.3%) on radiographs (P = .024). Overresected cams were detected in 31 patients (38.8%) on CT scans versus 14 patients (17.5%) on radiographs (P = .0049). Concomitant areas of under- and overresection were detected in 12 patients (15.0%) on CT scans versus 3 patients (3.8%) on radiographs (P = .027). Conclusion: CT scans with 3D planning software may be more sensitive than traditional radiographic views at detecting aberrant proximal femoral anatomy in the setting of failed FAI surgery. The use of 3D planning software may be considered as an adjunctive tool to better understand complex deformity in the proximal femur for the planning of revision hip arthroscopy.

Feature extraction and fusion using deep convolutional neural networks for Covid-19 detection using CT and X-RAY images

COVID-19 represents a novel variant of the coronavirus disease, having rapidly disseminated across the globe. In recent studies pertaining to computer vision, image processing, and classification techniques, numerous methodologies have been introduced employing chest X-ray images and computerized tomography (CT) images. This investigation introduces novel fused deep features founded on CT and X-ray image classification, with the aim of detecting COVID-19 disease. The proposed approach encompasses three key phases: deep feature generation, iterative feature selection, and classification. During the feature generation phase, well-known pre-trained deep convolutional neural networks, namely DenseNet201, MobileNetV2, ResNet18, ResNet50, ResNet101, VGG16, and VGG19, were leveraged. Each network model generated 1000 features, which were subsequently fused, culminating in the acquisition of a final feature vector comprising 7000 dimensions. In order to distill the most pertinent information, the ReliefF and iterative maximum relevance minimum redundancy (RFImRMR) feature selection techniques were employed in the generation of the ultimate feature vector. To evaluate the performance of the proposed method, publicly available datasets of CT images and X-ray images were employed. Notably, the suggested deep learning approach attained accuracy rates of 99.33% and 93.10% for COVID-19 detection using CT and X-ray images, respectively. These achieved results serve as compelling evidence substantiating the efficacy of the proposed fused deep features and RFImRMR-based COVID-19 detection.

Perceived Utility of Intracranial Pressure Monitoring in Traumatic Brain Injury: A Seattle International Brain Injury Consensus Conference Consensus-Based Analysis and Recommendations.

Intracranial pressure (ICP) monitoring is widely practiced, but the indications are incompletely developed, and guidelines are poorly followed. To study the monitoring practices of an established expert panel (the clinical working group from the Seattle International Brain Injury Consensus Conference effort) to examine the match between monitoring guidelines and their clinical decision-making and offer guidance for clinicians considering monitor insertion. We polled the 42 Seattle International Brain Injury Consensus Conference panel members' ICP monitoring decisions for virtual patients, using matrices of presenting signs (Glasgow Coma Scale [GCS] total or GCS motor, pupillary examination, and computed tomography diagnosis). Monitor insertion decisions were yes, no, or unsure (traffic light approach). We analyzed their responses for weighting of the presenting signs in decision-making using univariate regression. Heatmaps constructed from the choices of 41 panel members revealed wider ICP monitor use than predicted by guidelines. Clinical examination (GCS) was by far the most important characteristic and differed from guidelines in being nonlinear. The modified Marshall computed tomography classification was second and pupils third. We constructed a heatmap and listed the main clinical determinants representing 80% ICP monitor insertion consensus for our recommendations. Candidacy for ICP monitoring exceeds published indicators for monitor insertion, suggesting the clinical perception that the value of ICP data is greater than simply detecting and monitoring severe intracranial hypertension. Monitor insertion heatmaps are offered as potential guidance for ICP monitor insertion and to stimulate research into what actually drives monitor insertion in unconstrained, real-world conditions.

Development and validation of a recurrent prediction model for patients with unilateral chronic subdural hematoma without hematoma volumetric analysis

Approximately 10% of patients with chronic subdural hematoma (CSDH) undergo reoperation after initial surgery. This study aimed to develop a predictive model for the recurrence of unilateral CSDH at initial surgery without hematoma volumetric analysis. This single-center retrospective cohort study evaluated pre- and postoperative computed tomography (CT) images of patients with unilateral CSDH. The pre- and postoperative midline shift (MLS), residual hematoma thickness, and subdural cavity thickness (SCT) were measured. CT images were classified based on the internal architecture of the hematoma (homogenous, laminar, trabecular, separated, and gradation subtypes). Total 231 patients with unilateral CSDH underwent burr hole craniostomy. After receiver operating characteristic analysis, preoperative MLS and postoperative SCT showed better areas under the curve (AUCs) (0.684 and 0.756, respectively). According to the CT classification of preoperative hematomas, the recurrence rate was significantly higher in the separated/gradation group (18/97, 18.6%) than in the homogenous/laminar/trabecular group (10/134, 7.5%). Four-point score was derived from the multivariate model using the preoperative MLS, postoperative SCT, and CT classification. The AUC of this model was 0.796, and the recurrence rates at 0-4 points were 1.7%, 3.2%, 13.3%, 25.0%, and 35.7%, respectively. Pre- and postoperative CT findings without hematoma volumetric analysis may predict CSDH recurrence.

A New Computed Tomography Classification for Acute Mesenteric Ischaemia: More Than a “Gut Feeling”?

A New Computed Tomography Classification for Acute Mesenteric Ischaemia: More Than a “Gut Feeling”?

Evaluation of the correlation between multiple organ calcification on CT and disease severity in patients with TAFRO syndrome

PurposeThe purpose of this study was to investigate the incidence of multiple organ calcification and the correlation between multiple organ calcification and clinical severity in patients with thrombocytopenia, anasarca, fever, reticulin fibrosis, renal dysfunction, and organomegaly (TAFRO) syndrome.MethodsWe retrospectively identified 13 patients with TAFRO syndrome who were treated at our hospital between February 2019 and March 2021. Computed tomography (CT) images of TAFRO patients, which were acquired at admission and one month after admission, were evaluated. Additionally, clinical and laboratory data related to organ calcification and severity classification of TAFRO syndrome were investigated. The correlation between the presence of organ calcification on CT and TAFRO syndrome-severity classification was evaluated.ResultsOne month after admission, calcification of the myocardium, adrenal glands, gallbladder wall, pancreas, kidney, skeletal muscle, and skin were observed in 38%, 46%, 15%, 15%, 15%, 23%, and 15% of the thirteen patients, respectively. The occurrence rate of calcifications in the myocardium, adrenal glands, and skeletal muscle was significantly higher in patients with a grade 4 or higher clinical severity than in those with a level up to grade 3 (p = 0.001, p = 0.005, and p = 0.035, respectively).ConclusionsOur results suggest that the higher the clinical severity in patients with TAFRO syndrome, the higher is the frequency of calcification in the myocardium, adrenal glands, and skeletal muscle; therefore, the assessment of these organ calcifications on CT images may be useful in predicting the severity of TAFRO syndrome.

Prevalence and Risk Factors for Intensive Care Unit Delirium After Traumatic Brain Injury: A Retrospective Cohort Study.

Delirium remains understudied after traumatic brain injury (TBI). We sought to identify independent predictors of delirium among intensive care unit (ICU) patients with TBI. This single-center retrospective cohort study evaluated adult patients with TBI requiring ICU admission. Outcomes included delirium days within the first 14days, as assessed by the Confusion Assessment Method-ICU (CAM-ICU). Models were adjusted for age, sex, insurance, Marshall head computed tomography classification, presence of subarachnoid hemorrhage (SAH), Injury Severity Score (ISS), need for cardiopulmonary resuscitation, maximum admission Glasgow Coma motor score, glucose level, hemoglobin level, and pupil reactivity. Delirium prevalence was 60%, with a median duration of 4days (interquartile range: 2-8) among ICU patients with TBI (n = 2,664). Older age, higher ISS, maximum motor score < 6, Marshall class II-IV, and SAH were associated with risk of increased delirium duration (all p < 0.001). In this large cohort, ICU delirium after TBI affected three of five patients for a median duration of 4days. Age, general injury severity, motor score, and features of intracranial hemorrhage were predictive of more TBI-associated delirium days. Given the high prevalence of ICU delirium after TBIand itsimpact on hospitalization, further work is needed to understand the impact of delirium and TBI on outcomes and to determine whether delirium risk can be minimized.

CT classification model of pancreatic serous cystic neoplasm and mucinous cystic neoplasm based on deep transfer learning.

Pancreatic cancer is a highly lethal disease. The preoperative distinction between pancreatic serous cystic neoplasm (SCN) and mucinous cystic neoplasm (MCN) remains a clinical challenge. The goal of this study is to provide clinicians with supportive advice and avoid overtreatment by constructing a convolutional neural network (CNN) classifier to automatically identify pancreatic cancer using computed tomography (CT) images. We construct a CNN model using a dataset of 6,173 CT images obtained from 107 pathologically confirmed pancreatic cancer patients at Shanghai Changhai Hospital from January 2017 to February 2022. We divide CT slices into three categories namely, SCN, MCN, and no tumor, to train the DenseNet201-based CNN model with multi-head spatial attention mechanism (MSAM-DenseNet201). The attention module enhances the network's attention to local features and effectively improves the network performance. The trained model is applied to process all CT image slices and finally realize the two categories classification of MCN and SCN patients through a joint voting strategy. Using a 10-fold cross validation method, this new MSAM-DenseNet201 model achieves a classification accuracy of 92.52%, a precision of 92.16%, a sensitivity of 92.16%, and a specificity of 92.86%, respectively. This study demonstrates the feasibility of using a deep learning network or classification model to help diagnose MCN and SCN cases. This, the new method has great potential for developing new computer-aided diagnosis systems and applying in future clinical practice.

Lightweight Classification Network for Pulmonary Tuberculosis Based on CT Images

With the continuous development of medical informatics and digital diagnosis, the classification of tuberculosis cases from computed tomography (CT) images of the lung based on deep learning is an important guiding aid in clinical diagnosis and treatment. Due to its potential application in medical image classification, this task has received extensive research attention. Existing related neural network techniques are still challenging in terms of feature extraction of global contextual information of images and network complexity in achieving image classification. To address these issues, this paper proposes a lightweight medical image classification network based on a combination of Transformer and convolutional neural network (CNN) for the classification of tuberculosis cases from lung CT. The method mainly consists of a fusion of the CNN module and the Transformer module, exploiting the advantages of both in order to accomplish a more accurate classification task. On the one hand, the CNN branch supplements the Transformer branch with basic local feature information in the low level; on the other hand, in the middle and high levels of the model, the CNN branch can also provide the Transformer architecture with different On the other hand, in the middle and high levels of the model, the CNN branches can also provide different local and global feature information to the Transformer architecture to enhance the ability of the model to obtain feature information and improve the accuracy of image classification. A shortcut is used in each module of the network to solve the problem of poor model results due to gradient divergence and to optimise the effectiveness of TB classification. The proposed lightweight model can well solve the problem of long training time in the process of TB classification of lung CT and improve the speed of classification. The proposed method was validated on a computed tomography (CT) image dataset provided by the First Hospital of Lanzhou University. The experimental results show that the proposed lightweight classification network for tuberculosis based on CT medical images of lungs can fully extract the feature information of the input images and obtain high accuracy classification results.

Analysis of Deep Learning Techniques for Brain Tumour Classification from CT &amp; MRI Images

Brain tumour detection in an initialpoint is a critical step to saving human life. Computed Tomography (CT) and Magnetic Resonance Image (MRI) provide very detailed information about brain tumour tissues. So the segmentation of tumour region is possible from pancreatic CT and brain MRI. CT and MRI is a non-invasive technique and it does not produce any harmful radiation to the patient. The patient suspected of tumour undergoes radiological evaluation such that the area, location and grade of the tumour can be predicted from the CT and MRI analysis. This critical information helps the doctors to decide about further treatment like chemotherapy, surgery, or radiation. The diagnosis requires an accurate and very fast segmentation and classification of CT and MRI images. But nowadays radiologists are doing this task manually and it is a tedious and time-consuming procedure. Also, there is a chance of variation in the result from one expert to another. Here comes the significance of automatic segmentation and classification of tumour types with the help of computers.The proposed work aims to develop an efficient system that can detect pancreatic and brain tumour and can classify the pancreatic CT and brain MRI into normal, benign or malignant. This work can be categorized into two approaches. Thus the dataset prepared for this research work contains CT and MRI images.The first approach proposes traditional machine learning technologies to achieve the goal. Image pre-processing, feature extraction, segmentation and classification are the various steps of the traditional machine learning method. A detailed investigation is performed through various feature extraction techniques and classification techniques for pancreatic (CT) and brain MRI. Discrete Wavelet Transform (DWT) feature, Grey Level Co-occurrence Matrix (GLCM) feature, Gabor feature, Tamura features and Edge Orientation Histogram (EOH) features and their combinations are used for the extraction of CT and MRI features. Benign tumours are non-cancerous, but malignant tumours are cancerous. In the first approach, the Support Vector Machine (SVM) is the main classifier used for pancreatic CT and brain MRI classification as normal, benign or malignant.In this technology, a huge amount of data and machines with high computational capabilities like Graphic Processing Unit (GPU) are available. Thus the second approach of this paper is to exploit all these available resources to produce accurate results. In this part, deep learning, the latest fast growing technology introduced in 2015 is used for the classification of brain MRI. A Deep Convolutional Neural Networks (DCNN) model is proposed to perform the classification task efficiently. The CNN results are compared with the results of a simple neural network classifier. This method provides accurate and it shows that deep learning based classification outperforms traditional machine learning techniques which produce an accurate result only. This research work again concentrates on the Transfer Learning (TL) methods to classify pancreatic CT and brain MRI.

A Unified Multi-Phase CT Synthesis and Classification Framework for Kidney Cancer Diagnosis With Incomplete Data.

Multi-phase computed tomography (CT) is widely adopted for the diagnosis of kidney cancer due to the complementary information among phases. However, the complete set of multi-phase CT is often not available in practical clinical applications. In recent years, there have been some studies to generate the missing modality image from the available data. Nevertheless, the generated images are not guaranteed to be effective for the diagnosis task. In this paper, we propose a unified framework for kidney cancer diagnosis with incomplete multi-phase CT, which simultaneously recovers missing CT images and classifies cancer subtypes using the completed set of images. The advantage of our framework is that it encourages a synthesis model to explicitly learn to generate missing CT phases that are helpful for classifying cancer subtypes. We further incorporate lesion segmentation network into our framework to exploit lesion-level features for effective cancer classification in the whole CT volumes. The proposed framework is based on fully 3D convolutional neural networks to jointly optimize both synthesis and classification of 3D CT volumes. Extensive experiments on both in-house and external datasets demonstrate the effectiveness of our framework for the diagnosis with incomplete data compared with state-of-the-art baselines. In particular, cancer subtype classification using the completed CT data by our method achieves higher performance than the classification using the given incomplete data.

Association of the Presenting Glasgow Coma Scale in patients who requires ICU admission or operative intervention following traumatic brain injury with the Marshall computed tomography (CT) classification of traumatic brain injury

BACKGROUND Glasgow coma scale (GCS) and Marshall computed tomography classification of traumatic brain injury can predict the severity of the brain injury in patients following trauma. This study aims to analyse the association between two, in patients who required ICU admission or neurosurgical intervention following trauma.&#x0D; METHODOLOGY Retrospective study of 64 patients who underwent ICU admission or neurosurgical intervention following traumatic brain injury from September 2017 to December 2020 in Nepal Mediciti Hospital. Majority of the mild head injury where CT scan was not performed, discharged from the emergency or did not need ICU admission or admitted in ward for observation, severe polytrauma were excluded from the study. Glasgow coma Scale was categorized into mild (13-15), moderate (9-12) and severe (&lt;8). The Marshall CT scan Grade was dichotomized into (1-3) and (4-6).&#x0D; RESULTS Out 64 patients, majority were male 48 (84.4%), mean age 42.33 (SD±16.16). In admitted patients, 48.4 %( GCS&lt; 8), 39.1% (GCS 9-12), 12.5% (GCS 13-15).The higher marshal grade (4-6) was present in 93.54 %( &lt;8), 48% (9-12), 25% (13-15). There was significant association of the GCS with the Marshall CT scan grade (p=0.00).&#x0D; CONCLUSION There is significant association between the presenting GCS and Marshall CT Scan grade following TBI. The more severe patients with decreasing GCS have higher Marshall CT Scan grade in CT scan of the brain.