CT Scan Images Research Articles

An Observation and Analysis the role of Convolutional Neural Network towards Lung Cancer Prediction

Lung cancer is one of the most serious and prevalent diseases, causing many deaths each year. Though CT scan images are mostly used in the diagnosis of cancer, the assessment of scans is an error-prone and time-consuming task. Machine learning and AI-based models can identify and classify types of lung cancer quite accurately, which helps in the early-stage detection of lung cancer that can increase the survival rate. In this paper, Convolutional Neural Network is used to classify Adenocarcinoma, squamous cell carcinoma and normal case CT scan images from the Chest CT Scan Images Dataset using different combinations of hidden layers and parameters in CNN models. The proposed model was trained on 1000 CT Scan Images of cancerous and non-cancerous cells to find the best combination of parameters in CNN to predict lung cancer accurately. The proposed system recorded the highest accuracy of 92.79%. In addition to that, the paper addresses 192 observations made using the CNN model.

Experimental study on fracture propagation in anisotropy rock under cyclic hydraulic fracturing

Hydraulic fracturing is commonly used to enhance the hydraulic conductivity of geothermal, oil, and gas reservoirs, particularly those situated in enclosed and unconventional formations. Shale oil and gas reservoirs exhibit distinct characteristics in rock mechanics parameters, morphology, natural fracture geometry, and geological features. Structural anisotropy in materials generally arises from changes in grain type and size, weak layering, and variations in layering angles. This study aims to investigate the impact of two key factors, namely cyclic injection and structural anisotropy, on the hydraulic fracturing process. To achieve this goal, the phenomenon of hydraulic fatigue in rock was examined through various cyclic injection methods, and true-triaxial hydraulic fracturing tests were conducted on cubic samples composed of different materials and layers. To evaluate and analyze the fracture morphology of these samples, CT scan X-ray imaging technology was employed. The results revealed that samples subjected to cyclic injection exhibited an average breakdown pressure approximately 30% lower than those subjected to monotonic injection. Fracture geometry in the cyclic injection samples displayed a network pattern with multiple crack branches, while monotonic injection primarily resulted in a single-surface pattern. In samples with differences in horizontal stress, the anisotropy of the sample did not affect the fracture, and the single fracture surface extended in the direction of maximum stress. The energy generated by cyclic injection using the damage-controlled method (cycle duration: 16 s) was concentrated around the borehole, creating a complex fracture network with shorter crack lengths. Conversely, cyclic injection using the ramp signals method (cycle duration: eight seconds) produced more straightforward cracks with greater expansion and penetration. Examination of CT scan images of the fractured samples revealed average tortuosity and fracture volume fraction parameters of 1.117 and 0.027 for cyclic injection, respectively, in contrast to 1.026 and 0.008 for monotonic injection. Furthermore, the S-N damage model based on the results of true-triaxial hydraulic fracturing was introduced to describe rock damage behavior and the concept of fatigue under cyclic loads.

Genetic-efficient fine-tuning with layer pruning on multimodal Covid-19 medical imaging

Medical image analysis using multiple modalities refers to the process of analyzing and extracting information from more than one type of image in order to gain a comprehensive understanding of a given subject. To maximize the potential of multimodal data in improving and enhancing our understanding of the disease, sophisticated classification techniques must be developed as part of the integration process to classify meaningful information from different types of data. A pre-trained model, such as those trained on large datasets such as ImageNet, has learned rich representations that can be used for various downstream tasks. Fine-tuning a pre-trained model refers to the process of further developing the model using the knowledge and representations gained from a pre-existing dataset. In comparison to training a model from scratch, fine-tuning allows knowledge to be transferred from the pre-trained model to the target task, thus improving performance and efficiency. In evolutionary search, the genetic algorithm (GA) is an algorithm that emulates the process of natural selection and genetics. In this context, a population of candidate solutions is generated, fitness is evaluated and new candidate solutions are generated by applying genetic operations such as mutation and crossover. Considering the above characteristics, the present study presents an efficient architecture called Selective-COVIDNet for analyzing COVID-19 cases using a novel selective layer-pruning algorithm. To detect COVID-19 from multimodal data, the current study will use a genetic algorithm to fine-tune the performance of pre-trained models by adjusting specific layers selectively. Furthermore, the proposed approach provides flexibility in the depth of two deep learning architectures, VGG-16 and MobileNet-V2. The impact of freezing specific layers on fine-tuning performance was assessed using five different strategies, namely Random, Odd, Even, Half, and Full Freezing. Therefore, existing pre-trained models can be enhanced for Covid-19 tasks while minimizing their computational burden. For evaluating the effectiveness of the proposed framework, two multi-modal standard datasets are used, including CT-scan images and electrocardiogram (ECG) recordings of individuals with COVID-19. From the conducted experiments, it is found that the proposed framework can detect Covid-19 effectively with accuracy of 98.48% for MobileNet-V2 and 99.65% for VGG-16.

Adaptation in brain structure and respiratory and olfactory structures across environmental gradients in African and North American muroid rodents.

Morphometric studies of 3D micro CT-scanned images can provide insights into the evolution of the brain and sensory structures but such data are still scarce for the most diverse mammalian order of rodents. From reviewed and new data, we tested for convergence to extreme aridity and high elevation in the sensory and brain morphology of rodents, from morphometric data from micro-CT X-ray scans of 174 crania of 16 species of three distantly related African murid (soft-furred mice, Praomyini, laminate-toothed rats, Otomyini, and gerbils, Gerbillinae) clades and one North American cricetid (deer mice and white-footed mice, Peromyscus) clade. Recent studies demonstrated convergent evolution acting on the oval window area of the cochlea (enlarged in extremely arid-adapted species of Otomyini and Gerbillinae) and on endocranial volume (reduced in high elevation taxa of Otomyini and Peromyscus). However, contrary to our predictions, we did not find evidence of convergence in brain structure to aridity, or in the olfactory/respiratory system (turbinate bones) to high elevation. Brain structure differed, particularly in the petrosal lobules of the cerebellum and the olfactory bulbs, between Otomyini and Gerbillinae, with extreme arid-adapted species in each clade being highly divergent (not convergent) from other species in the same clade. We observed greater "packing" of the maxillary turbinate bones, which have important respiratory functions, in Peromyscus mice from high and low elevations compared to the high-elevation African Praomyini, but more complex patterns within Peromyscus, probably related to trade-offs in respiratory physiology and heat exchange in the nasal epithelium associated with high-elevation adaptation.

Expansion of the CT-scans image set based on the pretrained DCGAN for improving the performance of the CNN

Diagnosis of COVID-19 based on CT scans images using artificial intelligence can massively reduce medical manpower, time, and resources and reduce misdiagnosis rate. However, there are few relevant image sets publicly available at present, which is not conducive to the training of models such as classification. Transfer learning has been extensively considered in image classification, but less utilized in image generation. This work attempts to expand this dataset with discriminators using Deep Convolutional Generative Adversarial Network (DCGAN) models of Imagenet pre-training models and verify the significance of pre-training models for generating image veracity. A DCGAN model using the Imagenet pre-training model (i.e. resnet18) and a DCGAN network without the pre-train model generated 100 positive and 100 negative images to extend the original dataset, respectively. These images were classified using a Convolutional Neural Network (CNN) binary classifier, and the accuracy improved from 82.67% to 85.33% after adding the pre-training model. Furthermore, the comparison shown by the Gradcam visualization shows that the discriminator with the pre-training model can better capture the key details of the images. This experiment shows that even though Imagenet and lung images are totally uncorrelated, there are still some features that can be transferred. It also demonstrates that Imagenet pre-training models can improve the quality of GAN-generated images, extending the application of migration transfer learning.

LMNS-Net: Lightweight Multiscale Novel Semantic-Net deep learning approach used for automatic pancreas image segmentation in CT scan images

LMNS-Net: Lightweight Multiscale Novel Semantic-Net deep learning approach used for automatic pancreas image segmentation in CT scan images

Relation factor analysis for the short-term preservation of ipsilateral renal function after partial nephrectomy

Objectives: To analyze the factors relative to the short-term preservation of ipsilateral renal function after partial nephrectomy. Methods: The clinical data of 83 patients who were treated with partial nephrectomy from December 2014 to December 2019 in the Department of Urology, Sun Yat-sen University Cancer Center were retrospectively analyzed. There were 54 males and 29 females, aging (M (IQR)) 49 (17) years (range: 27 to 74 years). The ischemia time in operation was 25 (18) minutes (range: 10 to 67 minutes). Emission computed tomography scan and CT scan were performed before (within 1 month) and after (3 to 12 months) surgery. The volume of the ipsilateral and contralateral kidney was measured on the basis of preoperative and postoperative CT scans. The glomerular filtration rate (GFR) specifically in each kidney was estimated by emission computed tomography. Recovery from ischemia is determined by the formula: GFR preservation/volume saved×100%. Linear regression was used to explore the factors ralative to the short-term preservation of ipsilateral renal function after partial nephrectomy. Results: The GFR preservation of the ipsilateral kidney was 80.9 (25.2) % (range: 31.0% to 109.4%). The volume loss of the kidney resulted in a decrease of 12.0% (5.8 ml/(min×1.96 m2)) of GFR, while the ischemic injury resulted in a decrease of 6.5% (2.5 ml/(min×1.96 m2)) of GFR. The volume saved from the ipsilateral kidney was 87.1 (12.9) % (range: 27.0% to 131.7%). Recovery from ischemia was 93.5 (17.5) % (range:44.3% to 178.3%). In multivariate analysis, GFR preservation of the ipsilateral kidney was significantly correlated with the volume saved of the ipsilateral kidney (β=0.383, 95%CI: 0.144 to 0.622, P=0.002). It was not related to the ischemia time (β=0.046, 95%CI:-0.383 to 0.475, P=0.831). Conclusion: In the condition of limited ischemic time, in the short term ipsilateral renal function after partial nephrectomy is mainly determined by the loss of kidney volume, while ischemic injury only plays a minor role.

The advantages of general anesthesia subthalamic deep brain stimulation for Parkinson's disease in the enhanced recovery after surgery: A randomized clinical trial.

With advancements in imaging and microelectrode recording techniques, general anesthesia (GA) has emerged as an alternative option for Parkinson's disease (PD) patients undergoing subthalamic nucleus deep brain stimulation (STN-DBS). In this study, we compared the advantages and disadvantages of using GA and local anesthesia for STN-DBS in enhanced recovery after surgery (ERAS). Surgical outcomes of STN-DBS were evaluated using the unified PD rating scales (UPDRS). CT and magnetic resonance imaging scans are used to evaluate intracranial conditions. State-trait anxiety inventory and hospital anxiety and depression scale are used to evaluate patients' perioperative psychology. Anesthesia method does not significantly impact the accuracy of microelectrode placement or the improvement of postoperative symptoms. However, the local anesthesia group had a higher incidence of intracranial air, as well as higher rates of postoperative complications such as headache, dizziness, vomiting, and delirium. GA effectively alleviated preoperative anxiety and resulted in lower levels of perioperative anxiety and psychological stress compared to local anesthesia. Additionally, the GA group had shorter surgery duration, earlier ambulation, and a shorter average hospital stay. DBS under GA is safe and effective. Due to shorter surgical duration, reduced occurrence of perioperative complications, effective reduction of preoperative anxiety, and faster postoperative recovery, DBS under GA is better aligned with the concept of ERAS.

Pure tone audiometry is a new method for evaluating congenital malformation of the middle and outer ear (CMMOE)

Background: The preoperative evaluation of Congenital Malformation of the Middle and Outer Ear (CMMOE) is very important. Jahrsdoerfer score commonly used at present, based on CT scanning images of the temporal bone, is often unable to accurately evaluate deformity and hearing level. Aims/Objectives: To investigate and promote a straightforward and easily accessible assessment method, pure tone audiometry, for the evaluation of CMMOE. Material and Methods: A total of 223 cases (244 ears) CMMOE with hearing data were retrospectively analyzed. Among them, 180 cases (197 ears) underwent exploratory tympanoplasty with clear conditions: ossicle numbers in 136 cases (147 ears) and morphology in 128 cases (138 ears) and vestibular window development in 137 cases (146 ears), and CT scans of temporal bone in 113 cases (120 ears). 1). The correlation was analyzed between ossicle numbers, ossicle morphology, Jahrsdoerfer score groups and their corresponding Average Air-Conduction Threshold of pure tone (AACT) at 0.5–4 KHz. 2) The AACT difference is compared among the above groups respectively and between the developed and undeveloped groups of vestibular window at 0.5–4 KHz and each frequency of 0.125–8 KHz. Spearman method was used for correlation analysis (calculating coefficient r and p values). For the data followed a normal distribution, a one-way analysis of variance (ANOVA) and t-test were employed, otherwise, Kruskal Wallis multiple local rank coincidence test and Wilcoxon rank sum test were used. p <0 .05 was considered statistically significant. Results: 1) The correlation coefficients between the groups of ossicle number scores, ossicle morphology scores, Jahrsdoerfer scores and their corresponding AACT are r = −0.187 (p <0 .05), r = −0.073 (p >0 .05) and r = −0.079 (p > 0.05), respectively. 2) Comparison of AACT difference based on ossicle number or morphological scores and Jahrsdoerfer scores with p > 0.05 among all groups, respectively. The AACT difference between the developed and undeveloped vestibular window groups is 5.5 (63.5/69.0) dB HL(p < .05) at 0.5–4KHz, out of 0.125–8 KHz frequency 1, 2, 4 KHz were 5.7 (65.0/70.7) dB HL, 8.4 (60.7/69.1) dB HL and 2 (61.5/63.5) dB HL, respectively, all p < 0.05, the other frequencies with all p > 0.05. Conclusions and Significance: 1) Ossicle number was correlated with AACT, but not for ossicle morphology and Jahrsdoerfer scores. 2) There was no significant difference in AACT corresponding to ossicle number or morphology scores and Jahrsdoerfer scores groups, but the patients with undeveloped vestibular window had poorer hearing than those with developed ones. Therefore, the AACT can evaluate the development of ossicle and vestibular window, and more directly reflect the hearing level than Jahrsdoerfer score. Pure tone audiometry is simple, widely used, and easily accessible, which making it a new assessment method of CMMOE.

Liver Cancer Segmentation through Enhanced Feature Extraction and Mapping using Improved Transfer Learning Techniques

The largest solid organ in the body is the liver. Numerous other vital functions it carries out include removing impurities from the blood flow, controlling blood coagulation, and maintaining healthy blood sugar levels. All blood leaving the intestines and stomach is directed to the liver as its final destination. Liver illness may be brought on by infections, inherited diseases, cancer, too many harmful substances, or other conditions. Medical experts estimate that 1.6% of men and women in India could be diagnosed with liver cancer at some time in their life. The interpretation of liver CT scans often involves semi-manual or manual techniques; however these techniques are costly, time-consuming, subjective, and prone to error. To improve the detection of liver cancer, these issues have been addressed and a number of computer algorithms have been developed. To create a fully automated method for quickly extracting liver tumours from CT scan images.

Pediatric Patient with Omental Cyst: A Rare Case (CT scan is not Enough to Differentiate from Other Abdominal Cysts)

Background. Omental cysts are rare intra-abdominal tumors in children. It is challenging to diagnose since other big abdominal cysts are difficult to distinguish from in clinical and CT imaging. Aim. This case report aims to accurately and thoroughly determine the diagnosis based on the anamnesis, clinical and imaging examinations to choose the best treatment for the patient. Case Report. A 4-year-old girl reported having an enlarged belly in the past two months; the anamnesis provided the diagnosis. A mobile, 15 cm by 15 cm mass with dull percussion was discovered during clinical investigations in the top border. It was determined by anamnesis, physical examination, and imaging that the diagnosis was an ovarian cyst. Instead of an ovarian cyst, we found an omental cyst during the operation, thus we had to conduct an excision and omentectomy and postoperatively achieved favorable results. She was discharged three day after. Discussion. An abdominal CT scan showed an ovarian cyst. In laparotomy, we found a giant omental cyst, not originating from the ovaries, and did an omentectomy to excision the cyst. CT scan shows that the giant abdominal cyst is difficult to distinguish between omental, mesenteric, and ovarian cysts. Precise identification of stomach cysts is necessary. Also, doctors need to read the CT scan image more carefully. Conclusion. Giant omentum cysts are difficult to distinguish clinically and imaging from ovarian and mesenteric cysts. There is a need for a more detailed history, physical examination, and support, as well as a more thorough reading of the CT scan.

Clinical Characteristics and Outcomes of Elderly Patients with Stage I Diffuse Large B-Cell Lymphoma: A Study By Jiangsu Cooperative Lymphoma Group (JCLG)

Introduction Stage I diffuse large B-cell lymphoma (DLBCL) is defined as the involvement of a single lymph node or a group of adjacent nodes, or the presence of isolated extranodal lesions without nodal involvement. Generally, the prognosis of stage I DLBCL is excellent with 5-year overall survival (OS) over 90%. However, differences in disease characteristics and prognosis between elderly and younger patients with stage I DLBCL were not clear. Methods In this study, we conducted a retrospective data collection from 255 newly diagnosed patients with stage I DLBCL who were above 60 years old. The data was collected from 19 medical centers of of Jiangsu Cooperative Lymphoma Group (JCLG) located in Jiangsu Province, China. To enroll, patients had to undergo staging using either positron emission tomography/computed tomography (PET/CT) or contrast-enhanced CT scans of the chest, abdomen, and pelvis (C/A/P CT) with contrast, as well as bone marrow examination. Results The clinical characteristics of the 255 patients are presented in Table 1. The median age at presentation was 69 years, with a range of 61 to 92 years old. Among the 255 patients, 65.9% had at least one coexistent disease. A high Charlson Comorbidity Index (CCI), defined as ≥2, was found in 10.1% of patients. 63.9% had extranodal disease. The most common sites of extranodal involvement were the stomach (37.4%), intestine (19.0%), testes (11%), breast (7.4%), skin/soft tissue (5.5%), and sinus/nose (5.5%). According to the Hans algorithm, the non-GCB subtype accounted for 63.7% of patients and did not show a significant difference between the nodal and extranodal groups. None of the patients were diagnosed with double-hit lymphoma. Additionally, EBER was found to be positive in 3.7% of patients (5/134). The treatment approaches are outlined in Table 1. A total of 84.5% patients received the R-CHOP regimen as their primary treatment. The median number of R-CHOP courses administered was 6. Among the 204 patients with treatment evaluation records, 183 (89.7%) achieved complete remission (CR). With a median follow-up time of 30 months, 32 patients died during the follow-up period. Among them, 18 died from causes unrelated to lymphoma at a median age of 73 years. The 3-year progression-free survival (PFS) rate was 81.5% and the 3-year OS rate was 85.6%. In the univariate analysis, age ≥75 years (HR 3.30, P &amp;lt; 0·001) and CCI ≥2 (HR 2.92, P = 0·002) were significantly associated with worse PFS; age ≥75 years (HR 3·29, P = 0·001), CCI ≥2 (HR 2.57, P = 0·022) and non-GCB subtype (HR 1.79, P = 0·018) were significantly associated with worse OS. In multivariate analysis, age ≥75 years (HR 2.91, P = 0·001) and CCI ≥2 (HR 2.32, P = 0·02) remained independent risk indicators for worse PFS; age ≥75 years (HR 2.57, P = 0·015) and non-GCB subtype (HR 1.74, P = 0·025) were independent risk indicators for worse OS. None of the prognostic models including IPI, NCCN-IPI or stage-modified International Prognostic Index (sm-IPI) demonstrated statistical significance. However, by incorporating age≥75 and CCI≥2 into the sm-IPI, we could be able to provide a more accurate prediction of the prognosis for elderly patients with stage I DLBCL (Figure 1). Relapse occurred in 20 patients at a median time of 9 months. 91.7% (11/12) of patients with early relapse (defined as PFS less than 24 months) experienced recurrence in the same anatomical site as the primary disease. In contrast, only 25% (2/8) of patients with late relapse exhibited relapse in the initial site. Conclusion To the best of our knowledge, this is the largest retrospective study conducted specifically on stage I DLBCL in the Asian population during the rituximab era. Our findings suggest that elderly patients with stage I DLBCL have a higher non-lymphoma-related mortality rate and a higher likelihood of relapse in the same anatomical site for early relapses compared to late relapses. Integrating age ≥75 and CCI score into the sm-IPI will help to better assess prognosis.

Congenital duodenal web causing partial obstruction with recurrent vomiting and abdominal distention in a toddler boy: a case report

BackgroundThe duodenal web is a thin, elongated, web-like structure that is one of the factors contributing to duodenal obstruction. Only 100 cases have been reported in the literature. We present a 2.5-year-old cachectic Afghan child who did not have any overt signs and symptoms of intestinal obstruction, like recurrent vomiting, abdominal distention, and weight loss. The web was discovered near the intersection of the third and fourth portions, which is an uncommon location for the duodenal web. The late presentation of congenital duodenal web with partial obstruction is rare but well-known and has been reported in this case.Case presentationA 2.5-year-old cachectic Afghan child who had recurrent vomiting and experienced abdominal distention was brought to Maiwand Teaching Hospital from the Jabelsuraj region of Parwan province. The patient was suffering from unusual signs and symptoms like recurrent vomiting, abdominal distention, weight loss, and constipation. The diagnosis of these anomalies was established by a detailed history, clinical features, and abdominal CT scan. In the computerized tomography scanning (CT-Scan) image reported, there was a web with stenosis and partial obstruction in the distal aspect of the third-to-fourth portion of the duodenum. After preoperative stabilization, the child was taken for surgery. The abdomen was opened by a right upper abdominal transverse incision. After web resection and duodenoplasty, the patient was shifted to the recovery room in satisfactory condition. The child was allowed to feed after 8 days, which he tolerated well.ConclusionCongenital duodenal web with partial obstruction is typically observed in the second and third years of life. It is suspected in patients with recurrent vomiting, abdominal distention, weight loss, and constipation. Partial obstruction may not have an overt presentation, making it a challenging diagnosis for general practitioners. Abdomen X-ray and CT scan usually confirm the diagnosis, and successful surgical intervention is recommended.

Advanced lung tumor diagnosis using a 3D deep neural network based CAD system

In the realm of interdisciplinary research, Computer-Aided Diagnosis (CAD) emerges as a pivotal tool, aiding radiologists in comprehending lung computer tomography scans (both 2D and 3D) for enhanced prediction of lung malignancy and benign cases with heightened accuracy. The objective of clinical examination is to furnish medical experts with more effective scientific insights and treatment strategies. As conventional computer-aided design frameworks prove to be increasingly challenging and time-consuming, this study delves into the contemporary landscape of clinical image analysis in computer-aided diagnosis, with a specific focus on the architecture of the Convolutional Neural Network (CNN). Employing the Kaggle Data Science Bowl dataset, encompassing CT scan images of lung tumors, our proposed CNN architecture has exhibited remarkable performance. Notably, the CNN Net deep learning algorithm achieved a peak classification accuracy of 94%, coupled with a substantial sensitivity and specificity of 90.2%. The extensive 15-fold cross-validation procedure further highlighted the robustness of the CNN network, reporting a minimal log loss of 0.54. This outcome underscores the superior diagnostic capabilities of the CNN architecture in the context of lung tumor diagnosis. In essence, this paper not only contributes to advancements in radio diagnosis and precision medication but also provides a valuable addition to the field by presenting a meticulous figure for computer-aided design utilization.

Artificial Intelligence based Covid19 detection using CTScan and Chest X-Ray Images: Machine Learning and Deep Learning techniques

Artificial Intelligence based Covid19 detection using CTScan and Chest X-Ray Images: Machine Learning and Deep Learning techniques

OGBN P06 Co-location of pulmonary abnormalities on high resolution CT imaging and pulmonary microaspiration reflux scan: Suggests a causative relationship

Abstract Background The relationship between micro-aspiration of refluxate as a form of gastro-oesophageal reflux disease (GORD) and pulmonary diseases including bronchiectasis and lung fibrosis has been circumstantially suggested but not proven. Clarification of the role of GORD and microaspiration in pulmonary disease may enable targeted therapies. This study aims to determine whether there is a relationship between anatomical location of aspiration of radiological tracer on scintigraphic imaging and changes associated with pulmonary disease on high resolution computed tomographic imaging of the lung (HRCT-L). Methods Data were obtained from a prospectively populated database of patients with treatment resistant typical and atypical GORD and suspected reflux pulmonary microaspiration. Patients who underwent microaspiration studies and had concurrent HRCT-L for chest symptoms were included. Modified novel digital technique of reflux pulmonary microaspiration scintigraphy using a previously validated technique was performed to evaluate the possibility of significant GORD or extra-oesophageal symptoms, including typical symptoms such as heartburn and reflux and atypical symptoms such as throat symptoms, cough, globus, mucous, throat clearing, dysphonia and recurrent respiratory infections. HRCT-L was obtained in clinically indicated patients and compared with scintigraphic findings. Results Pulmonary reflux microaspiration scanning was performed in 121 patients (75% female, 25% male). The mean age was 67 years (range: 31 – 91 years). All patients demonstrated reflux microaspiration on scintigraphy in the supine delayed phase and all had intermittent or continuous full column gastro-oesophageal reflux. Lung parenchymal changes were apparent in 66 of 73 patients who underwent HRCT-L. Fibrotic changes were demonstrated in 51% of patients and bronchiectactic changes in 46%. Correlation at anatomical lobar levels with areas of aspiration of isotope on scintigraphic imaging was seen in 97% of fibrosis patients and 85% of bronchiectasis patients. Conclusions Modified reflux pulmonary microaspiration scintigraphy demonstrated the presence of pulmonary aspiration on delayed imaging. This is the first imaging technique with the ability to confirm the presence of lung aspiration of gastric content. Given the large proportion of cases showing lobar co-location of pulmonary abnormality on HRCT-L, this study indicates a strong relationship between aspiration of refluxate and parenchymal lung changes associated with pulmonary diseases such as IPF and bronchiectasis. There may be a significant role for modified technique scintigraphy to accurately diagnose the presence of reflux aspiration, allowing further research into pathogenesis and management of pulmonary disease.

Visual transformer and deep CNN prediction of high-risk COVID-19 infected patients using fusion of CT images and clinical data

BackgroundDespite the globally reducing hospitalization rates and the much lower risks of Covid-19 mortality, accurate diagnosis of the infection stage and prediction of outcomes are clinically of interest. Advanced current technology can facilitate automating the process and help identifying those who are at higher risks of developing severe illness. This work explores and represents deep-learning-based schemes for predicting clinical outcomes in Covid-19 infected patients, using Visual Transformer and Convolutional Neural Networks (CNNs), fed with 3D data fusion of CT scan images and patients’ clinical data.MethodsWe report on the efficiency of Video Swin Transformers and several CNN models fed with fusion datasets and CT scans only vs. a set of conventional classifiers fed with patients’ clinical data only. A relatively large clinical dataset from 380 Covid-19 diagnosed patients was used to train/test the models.ResultsResults show that the 3D Video Swin Transformers fed with the fusion datasets of 64 sectional CT scans + 67 clinical labels outperformed all other approaches for predicting outcomes in Covid-19-infected patients amongst all techniques (i.e., TPR = 0.95, FPR = 0.40, F0.5 score = 0.82, AUC = 0.77, Kappa = 0.6).ConclusionsWe demonstrate how the utility of our proposed novel 3D data fusion approach through concatenating CT scan images with patients’ clinical data can remarkably improve the performance of the models in predicting Covid-19 infection outcomes.SignificanceFindings indicate possibilities of predicting the severity of outcome using patients’ CT images and clinical data collected at the time of admission to hospital.

A case report of early detection of rare femoral shaft bone metastasis in a penis cancer patient

Background. Bone metastasis is very common in the advanced stage of numerous carcinomas. In penile carcinoma, lymph nodes metastasis is somehow common but it is very rare reported to be secondary from penile cancer. till the date, there are only few cases of penis carcinoma reported bone metastasis in literature worldwide.Case Presentation. Herein, We presented a 51-year-old Nepalese male with squamous cell carcinoma of penis. computed tomography (ct) scan of the patient revealed that there was carcinoma involving glans penis and precure with bilateral external &amp; internal inguinal lymphadenopathies. After then, the patient was under gone for partial penectomy and bilateral inguinal lymphadenectomy and complete 6-cycle chemotherapy. After one year of treatment, patient developed thigh pain and headache and he advised to have magnetic Resonance imaging (mRi) of brain, 99mTc-MDP whole body bone scan and ct scan of pelvis and thigh. The examination report reveals that there was a sclerotic change in vertex of skull bone and moderate 99mTc-MDP uptake in right proximal shaft of femur just below the neck d/d metastasis. The histopathological examination of the true cut biopsy taken from the lesion of the femur showed metastatic keratinizing squamous cell carcinoma which is rare case of femoral shaft bone metastasis secondary from penile carcinoma. Then patient was sent for surgical reconstruction of femur. Based on the case studies review femur shaft bone metastasis from penile cancer is extremely rare.Conclusion. The best of our knowledge; this is the first early detected bone metastases to shaft of the femur in a patient with penile cancer. early diagnosis helps to radical treatment as well as palliative treatment. surgery is the preferred option of the treatments, especially for metastatic foci in the long bones.

Optimization of Intracranial Hemorrhage Using CT Scan Images and Feature Extraction

Intracranial bleeding is among the most severe forms of brain stroke. The neurologic effects and artery rupture cause bleeding in the brain and the tissue around it. Haemorrhage is classified based on where the bleeding occurs on the brain. This paper depicts the application of multiple machine-learning approaches to separate CT scan images into normal and pathological categories. Separate analysis is conducted on the functionality of the features extracted from the various texturing approaches, such as the Grey Level Co-occurrence Matrix (GLCM) and Grey Level Run-Length Matrix (GRLM). Particle Swarm Optimisation (PSO) and K-Nearest Neighbours are used to choose relevant characteristics that increase the classification accuracy for feature extraction. The findings demonstrate that these texture features have excellent discrimination accuracy

EFFECTIVENESS EDGE DETECTION OPERATOR CANNY TO IMPROVE IMAGE QUALITY THORAX CT SCAN IN CASES COVID-19

Background: Thorax CT scan is one of the medical supports that contributes most importantly to the diagnosis, especially cases of COVID-19. The disadvantage that has a CT scan image is noise. When the noise is high then the Signal Noise Ratio (SNR) value produced is low. Canny operator edge detection technique is one solution to animate noise. Purpose: Analyzing differences in image quality and anatomical information on thorax CT scan images in COVID-19 patients before and after the application of canny operator edge detection techniques. Method: A quasi experimental research on thorax CT Scan images before and after the application of canny operator edge detection which amounted to 10 samples. Image assessment is done by measuring noise, SNR, and anatomical information. Differences in image quality (noise and SNR) are tested with paired T-tests. Anatomical information is tested with the Wilcoxon Signed Rank Test. Result: There are differences in image quality in thorax CT scan images in COVID-19 patients before and after the application of canny operator edge detection techniques, with p-value &lt;0.001. There are differences in the anatomical information of thorax CT scan images in COVID-19 patients before and after the application of canny operator edge detection technique with a p-value of 0.004. Conclusion: Edge detection operator canny techniques are able to lower noise values, improve SNR and improve image anatomy information thorax CT scan in COVID-19 patients.