Current Medical Imaging - Volume 20, Issue 1, 2024

To evaluate the usefulness of unenhanced CT added to the portal venous phase in the diagnostic accuracy of acute colonic diverticulitis/sigmoiditis.

Between January 1^st and December 31^st, 2020, all consecutive adult patients referred to the radiology department for clinical suspicion of acute colonic diverticulitis/sigmoiditis were retrospectively screened. To be included, patients must have undergone a CT with both unenhanced (UCT) and contrast-enhanced portal venous phase CT (CECT). CT examinations were assessed for features of diverticulitis, complications, differential diagnosis and incidental findings using UCT + CECT association, medical management, and follow-up as the reference. Radiation doses were recorded on our image archiving system and assessed.

Of the 114 patients included (mean age was 67±18 years; 60% were female), 46 had acute colonic diverticulitis/sigmoiditis. No diagnosis of sigmoiditis/diverticulitis, complication or differential diagnosis was missed with the CECT alone. Apart from diverticulitis, only one 2 mm meatal urinary microlithiasis was missed with no impact on patient management. The confidence level in diagnosis was not increased by UCT. The average DLP of CECT was 450 mGy.cm, and 382 mGy.cm for UCT. The use of a single-phase CECT acquisition allowed a reduction of 45.9% of the irradiation.

Unenhanced CT is not necessary for patients addressed with clinical suspicion of acute colonic diverticulitis/sigmoiditis, and CECT alone protocol must be used.

During the collection process, the prostate capsula is prone to introduce salt and pepper noise due to gastrointestinal peristalsis, which will affect the precision of subsequent object detection.

A cascade optimization scheme for image denoising based on image fusion was proposed to improve the peak signal-to-noise ratio (PSNR) and contour protection performance of heterogeneous medical images after image denoising.

Anisotropic diffusion fusion (ADF) was used to decompose the images denoised by adaptive median filter, non-local adaptive median filter and artificial neural network to generate the base layer and detail layer, which were fused by weighted average and Karhunen-Loeve Transform respectively. Finally, the image was reconstructed by linear superposition.

Compared with the traditional denoising method, the image denoised by this method has a higher PSNR while maintaining the image edge contour.

Using the denoised dataset for object detection, the detection precision of the obtained model is higher.

This study aims to comprehensively assess the characteristics of patent foramen ovale (PFO) in relation to Cryptogenic Strok (CS) by utilizing transesophageal echocardiography (TEE) and contrast transthoracic echocardiography (c-TTE) and to identify high-risk factors associated with PFO-related CS.

Transcatheter PFO closure has demonstrated its effectiveness in preventing PFO-related CS. Therefore, understanding the specific structural attributes of PFO associated with CS is imperative.

Enrollment comprised 113 test patients who experienced CS in conjunction with PFO and 117 control patients diagnosed with migraine with PFO but without a history of stroke. The characteristics of the PFO were observed by TEE and c-TTE. A comparative analysis was undertaken to assess the variations in PFO characteristics between the test patients and controls, and to uncover the independent factors relevant to CS.

The patients in the test group were older than the controls. Both the height and length of the PFO during Valsalva exhibited greater dimensions in the test group when contrasted with controls. Notably, the test group presented higher incidence rates of low-angle PFO (defined as an angle between the inferior vena cava (IVC) and PFO ≤ 10°) and atrial septal aneurysm (ASA) as contrasted with the control group. Right-to-left shunt (RLS) III during Valsalva demonstrated a significantly elevated occurrence within the test group as opposed to the controls. Conversely, RLS II during Valsalva exhibited a significantly higher frequency in the controls in contrast to the tests. No significant disparities were observed between the two groups with respect to RLS I during Valsalva and all grades of RLS at rest. Multivariate analysis revealed that the length of the PFO during Valsalva, the presence of ASA, RLS III during Valsalva and low-angle PFO were independent relevant factors associated with CS.

The length of the PFO tunnel, low-angle PFO, RLS III during Valsalva and the presence of ASA were independent risk factors for CS. The combined utilization of TEE and c-TTE may prove valuable in identifying PFO patients at a heightened risk of CS and in facilitating the screening process for transcatheter PFO closure.

Accurate preoperative judgment of lymph node (LN) metastasis is a critical step in creating a treatment strategy and evaluating prognosis in rectal cancer (RC) patients.

This study aimed to explore the value of T1 mapping and amide proton transfer weighted (APTw) imaging in predicting LN metastasis in patients with rectal cancer.

In a retrospective study, twenty-three patients with pathologically confirmed rectal adenocarcinoma who underwent MRI and surgery from August 2019 to August 2021 were selected. Then, 3.0T/MR sequences included conventional sequences (T1WI, T2WI, and DWI), APTw imaging, and T1 mapping. Patients were divided into LN metastasis (group A) and non-LN metastasis groups (group B). The intra-group correlation coefficient (ICC) was used to test the inter-observer consistency. Mann-Whitney U test was used to compare the differences between the two groups. Spearman correlation analysis was performed to evaluate the correlation between T1 and APT values. Logistic regression and receiver operating characteristic (ROC) curve analyses were performed to assess the differential performance of each parameter and their combination. The difference between AUCs was compared using the DeLong test.

The APT value in patients with LN metastasis was significantly higher than in those without LN metastasis group (P=0.020). Also, similar results were observed for the T1 values (P=0.001). The area under the ROC curve of the APT value in the prediction of LN metastasis was 0.794; when the cutoff value was 1.73%, the sensitivity and specificity were 71.4% and 88.9%, respectively. The area under the ROC curve of the T1 value was 0.913; when the cutoff value was 1367.36 ms, the sensitivity and specificity were 78.6% and 100.0%, respectively. The area under the ROC curve of T1+APT was 0.929, with a sensitivity of 78.6% and specificity of 100.0%.

APT and T1 values show great diagnostic efficiency in predicting LN metastasis in rectal cancer.

Despite the availability of vaccines, there is an increasing number of SARS-CoV-2-breakthrough-infections.

The aim of this study was to determine whether there is a radiological difference in lung parenchymal involvement between infected vaccinated and unvaccinated patients. Additionally, we aimed to investigate whether vaccination has an impact on the course of illness and the need for intensive care.

This study includes all patients undergoing chest computed tomography (CT) or x-ray imaging in case of a proven SARS-CoV-2 infection between September and November 2021. Anonymized CT and x-ray images were reviewed retrospectively and in consensus by two radiologists, applying an internal severity score scheme for CT and x-ray as well as CARE and BRIXIA scores for x-ray. Radiological findings were compared to vaccination status, comorbidities, inpatient course of the patient’s illness and the subjective onset of symptoms.

In total, 38 patients with acute SARS-CoV-2 infection underwent a CT scan, and 168 patients underwent an x-ray examination during the study period. Of these, 32% were vaccinated in the CT group, and 45% in the x-ray group. For the latter, vaccinated patients exhibited significantly more comorbidities (cardiovascular (p=0.002), haemato-oncological diseases (p=0.016), immunosuppression (p=0.004)), and a higher age (p<0.001). Vaccinated groups showed significantly lower extent of lung involvement (severity scores in CT cohort and x-ray cohort both p≤0.020; ARDS 42% in unvaccinated CT cohort vs. 8% in vaccinated CT cohort). Furthermore, vaccinated patients in the CT cohort had significantly less need for intensive care treatment (p=0.040).

Our data suggest that vaccination, in the case of breakthrough infection, favours a milder course of illness concerning lung parenchymal involvement and the need for intensive care, despite negative predictors, such as immunosuppression or other pre-existing conditions.

Empirical curvelet and ridgelet image fusion is an emerging technique in the field of image processing that aims to combine the benefits of both transforms.

The proposed method begins by decomposing the input images into curvelet and ridgelet coefficients using respective transform algorithms for Computerized Tomography (CT) and magnetic Resonance Imaging (MR) brain images.

An empirical coefficient selection strategy is then employed to identify the most significant coefficients from both domains based on their magnitude and directionality. These selected coefficients are coalesced using a fusion rule to generate a fused coefficient map. To reconstruct the image, an inverse curvelet and ridgelet transform was applied to the fused coefficient map, resulting in a high-resolution fused image that incorporates the salient features from both input images.

The experimental outcomes on real-world datasets show how the suggested strategy preserves crucial information, improves image quality, and outperforms more conventional fusion techniques. For CT Ridgelet-MR Curvelet and CT Curvelet-MR Ridgelet, the authors' maximum PSNRs were 58.97 dB and 55.03 dB, respectively. Other datasets are compared with the suggested methodology.

The proposed method's ability to capture fine details, handle complex geometries, and provide an improved trade-off between spatial and spectral information makes it a valuable tool for image fusion tasks.

An anomalous left coronary artery from the pulmonary artery (ALCAPA) is a rare heart malformation, with 90% of patients dying during the first year of life. If the right coronary artery compensation and multiple collateral circulation are sufficiently established, the patient's myocardial ischemia symptoms are mild and appear later, which is called the adult type ALCAPA.

A 42-year-old woman presented to our hospital with one-month history of the aggravation of active shortness of breath which gradually progressed to nocturnal paroxysmal shortness of breath and cough. Admission physical examination suggested mild edema of both lower limbs. Transthoracic echocardiography (TTE) showed that a small vessel shadow was abnormally connected to the pulmonary artery (PA), and moderate pulmonary artery hypertension. Coronary computed tomography angiography (CTA) showed an anomalous origin of the left main coronary artery (LMCA) dividing into the left anterior descending (LAD) and left circumflex (LCX) artery from the PA, with no clear connection to the left coronary sinus. The right coronary artery (RCA) was significantly dilated and originated from the normal Valsalva sinus. It was accompanied by multiple collateral circulations, most of which traveled anterior to the right ventricular free wall and anterior interventricular sulcus, and some emanated from the posterior descending branch of the posterior interventricular sulcus and walked toward the posterolateral wall of the left ventricle.

Coronary computed tomography angiography (CTA) can be used to visualize the abnormal origin and distribution of the coronary artery's course and may be the first choice in the diagnosis of ALCAPA.

A clinical medical image provides vital information about a person's health and bodily condition. Typically, doctors monitor and examine several types of medical images individually to gather supplementary information for illness diagnosis and treatment. As it is arduous to analyze and diagnose from a single image, multi-modality images have been shown to enhance the precision of diagnosis and evaluation of medical conditions.

Several conventional image fusion techniques strengthen the consistency of the information by combining varied image observations; nevertheless, the drawback of these techniques in retaining all crucial elements of the original images can have a negative impact on the accuracy of clinical diagnoses. This research develops an improved image fusion technique based on fine-grained saliency and an anisotropic diffusion filter to preserve structural and detailed information of the individual image.

In contrast to prior efforts, the saliency method is not executed using a pyramidal decomposition, but rather an integral image on the original scale is used to obtain features of superior quality. Furthermore, an anisotropic diffusion filter is utilized for the decomposition of the original source images into a base layer and a detail layer. The proposed algorithm's performance is then contrasted to those of cutting-edge image fusion algorithms.

The proposed approach cannot only cope with the fusion of medical images well, both subjectively and objectively, according to the results obtained, but also has high computational efficiency.

Furthermore, it provides a roadmap for the direction of future research.

Exploring the relationship between the signal-to-noise ratio (SNR) of organs and size-specific dose estimate (SSDE) in tube current modulation (TCM) chest CT examination.

Forty patients who received TCM chest CT scanning were retrospectively collected and divided into four groups according to the tube voltage and sexes. We chose to set up the region of interest (ROI) at the tracheal bifurcation and its upper and lower parts in slice images of the heart, aorta, lungs, paracranial muscles, and female breast, and the SNR of each organ was calculated. We also calculated the corresponding axial volume CT dose index (CTDIvolz) and axial size-specific dose estimate (SSDEz).

The correlation analysis showed that the correlation between the SNR of the slice images of most organs and SSDEz was more significant than 0.8, and that between the SNR and CTDIvol was more significant than 0.7. The simple linear regression analysis results showed that when the sex is the same, the SNR of the same organ at 100kVp was higher than 120kVp, except for the lung. In multiple regression analysis, the result indicated that the determination coefficients of the SNR and SSDEz of the four groups were 0.934, 0.971, 0.905, and 0.709, respectively.

In chest CT examinations with TCM, the correlation between the SNR of each organ in slice images and SSDEz was better than that of CTDIvolz. And when the SSDEz was the same, the SNR at 100 kVp was better than that at 120 kVp.

Segmenting tumors in MRI scans is a difficult and time-consuming task for radiologists. This is because tumors come in different shapes, sizes, and textures, making them hard to identify visually.

This study proposes a new method called the enhanced regularized ensemble encoder-decoder network (EREEDN) for more accurate brain tumor segmentation.

The EREEDN model first preprocesses the MRI data by normalizing the intensity levels. It then uses a series of autoencoder networks to segment the tumor. These autoencoder networks are trained using back-propagation and gradient descent. To prevent overfitting, the EREEDN model also uses L2 regularization and dropout mechanisms.

The EREEDN model was evaluated on the BraTS 2020 dataset. It achieved high performance on various metrics, including accuracy, sensitivity, specificity, and dice coefficient score. The EREEDN model outperformed other methods on the BraTS 2020 dataset.

The EREEDN model is a promising new method for brain tumor segmentation. It is more accurate and efficient than previous methods. Future studies will focus on improving the performance of the EREEDN model on complex tumors.

Stroke is reportedly the biggest cause of death and disability in the world, according to the World Health Organisation (WHO). The severity of a stroke can be lowered by recognising the many stroke warning indicators early on. Using CNN, the primary goal of this study is to predict the likelihood that a brain stroke would develop at an early stage.

The novelty of the proposed work is to acquire models that can accurately differentiate between stroke and no-stroke (normal) cases using MR Imaging sequences like DWI, SWI, GRE and T2 FLAIR aiding in timely diagnosis and treatment decisions.

A dataset comprising real time MRI scans of patients with stroke and no-stroke conditions was collected and preprocessed for model training. The preprocessing involves standardizing the resolution of the images, normalizing pixel values, and augmenting the dataset to enhance model generalization. The ResNet, DenseNet, EfficientNet, and VGG16 architectures were implemented and trained on the preprocessed dataset. The training process involved optimizing model parameters using stochastic gradient descent and minimizing the loss function.

The results demonstrate promising performance across all models by obtaining an accuracy of 98% for ResNet, DenseNet and EfficientNet, while 97% for VGG16 in differentiating the stroke using real time MRI data.

The proposed work explored the effectiveness of CNN models, including ResNet, DenseNet, EfficientNet, and VGG16, for the differentiation of stroke and no-stroke cases. The models were trained and evaluated using a real-time dataset of brain MR Images. The obtained accuracies highlight the potential of CNN models in accurately differentiating between stroke and non-stroke cases.

The novel coronavirus pandemic has caused a global health crisis, placing immense strain on healthcare systems worldwide. Chest X-ray technology has emerged as a critical tool for the diagnosis and treatment of COVID-19. However, the manual interpretation of chest X-ray films has proven to be inefficient and time-consuming, necessitating the development of an automated classification system.

In response to the challenges posed by the COVID-19 pandemic, we aimed to develop a deep learning model that accurately classifies chest X-ray images, specifically focusing on lung regions, to enhance the efficiency and accuracy of COVID-19 and pneumonia diagnosis.

We have proposed a novel deep network called “FocusNet” for precise segmentation of lung regions in chest radiographs. This segmentation allows for the accurate extraction of lung contours from chest X-ray images, which are then input into the classification network, ResNet18. By training the model on these segmented lung datasets, we sought to improve the accuracy of classification.

The performance of our proposed system was evaluated on three types of lung regions in normal individuals, COVID-19 patients, and those with pneumonia. The average accuracy of the segmentation model (FocusNet) in segmenting lung regions was found to be above 90%. After re-classification of the segmented lung images, the specificities and sensitivities for normal, COVID-19, and pneumonia were excellent, with values of 98.00%, 99.00%, 99.50%, and 98.50%, 100.00%, and 99.00%, respectively. ResNet18 achieved impressive area under the curve (AUC) values of 0.99, 1.00, and 0.99 for classifying normal, COVID-19, and pneumonia, respectively, on the segmented lung datasets. Moreover, the AUC values of the three groups increased by 0.02, 0.02, and 0.06, respectively, when compared to the direct classification of unsegmented original images. Overall, the accuracy of lung region classification after processing the datasets was 99.3%.

Our deep learning-based automated chest X-ray classification system, incorporating lung region segmentation using FocusNet and subsequent classification with ResNet18, has significantly improved the accuracy of diagnosing respiratory lung diseases, including COVID-19. The proposed approach has great potential to revolutionize the diagnosis of COVID-19 and other respiratory lung diseases, offering a valuable tool to support healthcare professionals during health crises.

Immunoglobulin G4 (Ig G4)-related disease is rare; however, it is a fibroinflammatory disease that has been studied a lot so far. Although the expression pattern varies depending on the organ affected, it usually manifests as organ hypertrophy and organ dysfunction.

A 46-year-old man was referred to our otorhinolaryngology department for left submandibular swelling and tenderness that occurred 2 weeks ago. He was treated with antibiotics (augmentin 625mg, per oral) for 2 weeks, but his symptoms did not improve, and his white blood cell (WBC) count was 10,500 /μL (normal 3,800−10,000 /μL).

A mass-like lesion of the submandibular space has been concluded and the laboratory findings have been satisfactory (IgG4 level); IgG4-related disease, which is rare, but recently often reported, can be included in the differential diagnosis.

The purpose of this study was to assess the diagnostic value of magnetic resonance imaging (MRI) in staging and treatment of cervical cancer in pregnancy, and to evaluate the benefit of apparent diffusion coefficient (ADC) during neoadjuvant chemotherapy management.

This was a retrospective cohort study. Patients were divided into two groups according to the stage of cervical cancer. The mean term of pregnancy at the time of the diagnosis was the early second trimester (range 10-27 weeks) and the median age was 33 years (range 26-40 years). The abdominal and pelvic MRI images and clinical data of these patients were reviewed. Tumor size, local tumor spread, and nodal involvement were evaluated using an MRI dataset. The treatment and follow-up imaging were analyzed as well, and the ADC was measured before and after the chemotherapy.

16 patients with histopathologically confirmed cervical cancer during pregnancy were retrospectively enrolled. 7 patients were diagnosed with local cervical cancer (FIGO stage IAI) and designated as early stage group, as the lesion was invisible on MRI. In this group, pregnancies were allowed to continue until cesarean delivery (CD) at 38-41 weeks. The other 9 patients presenting with local or extensive cervical cancer (FIGO stage IB2-IIA2) were designated as the advanced-stage group. The lesion could be measured and analyzed on MRI. They were treated with neo-adjuvant chemotherapy in pregnancy. Among them, 6 patients underwent TP regimen (paclitaxel 135~175 mg/m2 plus cisplatin 70~75 mg/m2), while 3 patients received TC regimen (paclitaxel 135~175 mg/m2 plus carboplatin AUC=5). NACT was performed for 1 to 2 courses before surgery. ADC demonstrated significant differences before and after chemotherapy administered during pregnancy (1.06 ± 0.12 sec/mm2 vs. 1.34 ± 0.21 sec/mm2).

MRI has been found to be helpful in staging cervical cancer in pregnancy. Patients with stage IA confirmed by MRI can choose conservative treatment and continue the pregnancy until term birth. MRI can dynamically monitor the efficacy of chemotherapy for patients with stage IB and above during pregnancy. ADC value can have a potential role in the evaluation of chemotherapy efficacy.

Diabetic Retinopathy (DR) is a growing problem in Asian countries. DR accounts for 5% to 7% of all blindness in the entire area. In India, the record of DR-affected patients will reach around 79.4 million by 2030.

The main objective of the investigation is to utilize 2-D colored fundus retina scans to determine if an individual possesses DR or not. In this regard, Engineering-based techniques such as deep learning and neural networks play a methodical role in fighting against this fatal disease.

In this research work, a Computational Model for detecting DR using Convolutional Neural Network (DRCNN) is proposed. This method contrasts the fundus retina scans of the DR-afflicted eye with the usual human eyes. Using CNN and layers like Conv2D, Pooling, Dense, Flatten, and Dropout, the model aids in comprehending the scan's curve and color-based features. For training and error reduction, the Visual Geometry Group (VGG-16) model and Adaptive Moment Estimation Optimizer are utilized.

The variations in a dataset like 50%, 60%, 70%, 80%, and 90% images are reserved for the training phase, and the rest images are reserved for the testing phase. In the proposed model, the VGG-16 model comprises 138M parameters. The accuracy is achieved maximum rate of 90% when the training dataset is reserved at 80%. The model was validated using other datasets.

The suggested contribution to research determines conclusively whether the provided OCT scan utilizes an effective method for detecting DR-affected individuals within just a few moments.

To evaluate the predictive value of 3.0T MRI Intravoxel Incoherent motion diffusion-weighted magnetic resonance imaging (IVIM-DWI) combined with texture analysis (TA) in the histological grade of rectal adenocarcinoma.

Seventy-one patients with rectal adenocarcinoma confirmed by pathology after surgical resection were collected retrospectively. According to pathology, they were divided into a poorly differentiated group (n=23) and a moderately differentiated group (n=48). The IVIM-DWI parameters and TA characteristics of the two groups were compared, and a prediction model was constructed by multivariate logistic regression analysis. ROC curves were plotted for each individual and combined parameter.

There were statistically significant differences in D and D* values ​​between the two groups (P < 0.05). The three texture parameters SmallAreaEmphasis, Median, and Maximum had statistically significant differences between groups (P = 0.01, 0.004, 0.009, respectively). The logistic regression prediction model showed that D*, the median, and the maximum value were significant independent predictors, and the AUC of the regression prediction model was 0.860, which was significantly higher than other single parameters.

3.0T MRI IVIM-DWI parameters combined with TA can provide valuable information for predicting the histological grades of rectal adenocarcinoma one week before the operation.

With the development of vascular intervention, pseudoaneurysm complications are increasing. Ultrasound-guided thrombin injection (UGTI) is currently the treatment of choice for pseudoaneurysm, but the pharmacological properties of thrombin may trigger acute thrombosis within the vessel lumen. Despite a very low incidence, this type of primary arterial thrombosis is a serious complication of UGTI, and cases involving multiple branches of the lower limb arteries are particularly rare.

Here, we report a case of a 65-year-old male who underwent UGTI for the treatment of an iatrogenic pseudoaneurysm of the femoral artery complicated by acute thrombosis of multiple arteries in the lower limbs, and the patient ultimately underwent a successful thrombectomy.

We reviewed the case and analyzed the possible etiologic causes, providing a reference for future clinical work.

Pulmonary embolism (PE) is a relatively rare vascular complication of acute pancreatitis (AP), and its mortality rate is high. To our knowledge, relevant literature reports still need to be summarized. In this study, we analyzed the clinical characteristics, treatment, and prognosis of five patients with AP complicated by PE and summarized and reviewed the relevant literature.

Clinical data of patients with AP complicated by PE treated in Taizhou Hospital of Zhejiang Province between January 2017 and September 2022 were retrospectively collected. Combined with the relevant literature, the clinical characteristics, treatment, and prognoses of patients with AP combined with PE were analyzed and summarized.

Five patients were eventually enrolled in this study. Among the five patients with AP complicated by PE, all (100%) had symptoms of malaise, primarily chest tightness, shortness of breath, and dyspnea. All patients (100%) had varied degrees of elevated D-dimer levels and a significant decrease in the pressure of partial oxygen (PO2) and pressure of arterial oxygen to fractional inspired oxygen concentration ratio (PaO2/FiO2). Computed tomographic angiography (CTA) or pulmonary ventilation/perfusion imaging revealed a pulmonary artery filling defect in these patients. One patient (20%) had left calf muscular venous thrombosis before the occurrence of PE. Four patients (80%) were treated with low-molecular-weight heparin (LMWH), and one patient (20%) was treated with rivaroxaban during hospitalization; all continued oral anticoagulant therapy after discharge. All patients (100%) were cured and discharged. No patients showed recurrence of AP or PE.

PE is a rare but life-threatening complication of AP. However, once diagnosed, early treatment with anticoagulation or radiological interventional procedures is effective, and the prognosis is good.

Pulmonary embolism (PE) is a rare but life-threatening complication of acute pancreatitis (AP). Its early diagnosis and timely anticoagulation or radiological intervention can reduce mortality. However, only nine cases have been reported in the English literature thus far, and they are all case reports. Our study is the first systematic analysis of patients with AP combined with PE with a review of the relevant literature. Our patients and those reported in the literature were discharged with good prognoses under treatment such as anticoagulation and vascular intervention. These cases remind clinicians that, in patients with AP, especially those with risk factors for venous thrombosis, it is necessary to monitor the D-dimer level dynamically. Clinicians should pay attention to AP patients' symptoms and related examinations to reduce the chance of a missed diagnosis or misdiagnosis of PE.