Computed Tomography Research Articles

Integrating 3D technology with the Sampaio classification for enhanced percutaneous nephrolithotomy in complex renal calculi treatment

BackgroundTo investigate the safety and efficacy of percutaneous nephrolithotomy (PCNL) in the treatment of complicated renal calculi by integrating three-dimensional (3D) computed tomography (CT) reconstruction with the Sampaio classification of the renal collecting system.MethodsSixty-four consecutive patients with complex kidney calculi who underwent PCNL between January 2019 and October 2023 were retrospectively analyzed and divided into experimental group (3D printing) and control group (CT imaging) according to their willingness to pay for 3D imaging. Both groups underwent preoperative CT urography. The Digital Imaging and Communications (DICOM) in Medicine data of the experimental group from CT imaging were used for 3D reconstruction and model printing. Then, the Sampaio classification system was used to design the puncture channel and develop a surgical strategy.ResultsThe 3D-printed models of the experimental group successfully displayed the Sampaio classification system. There was no significant difference in the baseline parameters between the groups. Compared with the control group, the experimental group exhibited significant improvements in the puncture time, number of puncture needles, number of puncture channels, target calyx consistency, number of first puncture channels, and stone clearance. There were no significant differences in the total operative time, decrease in the hemoglobin level, length of hospital stay, and postoperative complications between the groups.ConclusionsIntegration of 3D technology with the Sampaio classification of the renal collecting system can enhance the preoperative evaluation and planning of percutaneous renal access. This approach allows a more precise method of PCNL for treating complex renal calculi.

MaskAppendix: Backbone-Enriched Mask R-CNN Based on Grad-CAM for Automatic Appendix Segmentation

Background: A leading cause of emergency abdominal surgery, appendicitis is a common condition affecting millions of people worldwide. Automatic and accurate segmentation of the appendix from medical imaging is a challenging task, due to its small size, variability in shape, and proximity to other anatomical structures. Methods: In this study, we propose a backbone-enriched Mask R-CNN architecture (MaskAppendix) on the Detectron platform, enhanced with Gradient-weighted Class Activation Mapping (Grad-CAM), for precise appendix segmentation on computed tomography (CT) scans. In the proposed MaskAppendix deep learning model, ResNet101 network is used as the backbone. By integrating Grad-CAM into the MaskAppendix network, our model improves feature localization, allowing it to better capture subtle variations in appendix morphology. Results: We conduct extensive experiments on a dataset of abdominal CT scans, demonstrating that our method achieves state-of-the-art performance in appendix segmentation, outperforming traditional segmentation techniques in terms of both accuracy and robustness. In the automatic segmentation of the appendix region in CT slices, a DSC score of 87.17% was achieved with the proposed approach, and the results obtained have the potential to improve clinical diagnostic accuracy. Conclusions: This framework provides an effective tool for aiding clinicians in the diagnosis of appendicitis and other related conditions, reducing the potential for diagnostic errors and enhancing clinical workflow efficiency.

Geomechanical aspects of stabilizing arsenic trioxide roaster waste in cemented paste backfill at the Giant Mine, Canada

Giant Mine, an abandoned gold mine near Yellowknife in the Northwest Territories of Canada, generated significant arsenic trioxide roaster waste (ATRW) during its operations, posing a substantial environmental hazard. This study explored the feasibility of stabilizing ATRW by incorporating it into cemented paste backfill (CPB). Using response surface methodology (RSM), CPB samples with varying mixing ratios were analyzed to identify key parameters influencing strength. Unconfined compressive strength (UCS) tests assessed physical stability, while saturated hydraulic conductivity and computed tomography (CT) analyses examined the microstructure of the CPB. The results revealed that CPB samples prepared with general use (GU) cement exhibited significantly higher strength than those with a GU and lime kiln dust (LKD) mixture. Binder and solid contents were identified as the most critical factors influencing UCS, with binder content having a more pronounced influence. Curing time was found to be non-significant. Higher binder and solid contents correlated with higher UCS values in the CPB samples. The addition of 10% wt. ATRW reduced the UCS by over 30%, particularly in samples with lower binder and solid contents. Although microstructure differences were not evident in saturated hydraulic conductivity tests, CT scans showed increased formation of high-density arsenic-containing materials in samples with the highest UCS, especially those using GU binder. These findings suggest that optimizing binder and solid contents is crucial for enhancing CPB strength and effectively stabilizing ATRW.

A month of diagnostic imaging studies in an acute supportive/palliative care unit.

The aim was to assess the characteristics of patients who required imaging studies during admission to an acute supportive palliative care unit (ASPCU). A consecutive number of patients who performed imaging studies during ASPCU admission in a month period were assessed. Epidemiological data, ongoing anticancer treatment, cancer diagnosis, reasons for admission, referral, and type of imaging study were recorded. Indications, findings, consequent actions for treatment, prognosis, and discharge were also collected. Twenty-one of 56 patients admitted to ASPCU in the period taken into consideration were assessed. Pain and deterioration of the general condition were the most frequent indications for admission. Computed tomography (CT) was the most frequent imaging study performed. Indications for performing imaging studies depended on individual clinical needs. Findings suggested different clinical decisions, after a comprehensive oncological and palliative care assessment and family conference, the most frequent of which was to withdraw oncological treatments. The majority of patients underwent transition to palliative care and were discharged home or to hospice. Imaging studies were of paramount importance for clinical treatment and decision-making process in an intensive ASPCU. There is the need to explore the need and the possible outcomes of imaging studies, as well cost-effectiveness in any ASPCU.

Recommendation for Clinical T Staging in Patients with Non-Small Cell Lung Cancer: Volumetric Measurement: A Retrospective Study from Turkey.

Currently, clinical T staging in non-small cell lung cancer (NSCLC) is based on the largest radiological diameter observed on computed tomography (CT). Under this system, tumors with varying shapes-such as spherical, amorphous, or spiculated tumors- can be assigned the same T stage even with different volumes. We aimed to propose a 3-dimensional (3D) volumetric staging system for NSCLC as an alternative to diameter-based T staging and to conduct comparative survival analyses between these methods. We retrospectively analyzed data from patients who underwent surgery for pT1-4N0M0 primary NSCLC between January 2018 and May 2022. Digital Imaging and Communications in Medicine data from patient CT scans were uploaded to 3D Slicer software for volumetric tumor measurement. Using the paired samples t-test or the Wilcoxon test, we compared the expected tumor volumes, calculated by tumor diameter, with the actual volumes measured by 3D Slicer. Receiver operating characteristic analysis was employed to determine the cut-off value for tumor volume. Kaplan-Meier analysis was utilized to assess overall survival, while the log-rank method was applied to compare survival differences between groups. The significance of changes in T stage was evaluated using the marginal homogeneity test. The study included 136 patients. Significant differences were observed between expected and actual tumor volumes (p=0.01), and associated changes in T stage were also significant (p=0.04). The survival analysis performed using tumor volume (p=0.009) yielded superior results compared to that based on diameter (p=0.04) in paients with early tumor stage. T-factor staging based on tumor volume could represent an alternative staging method for NSCLC.

Association Between Body Composition Measured by Artificial Intelligence and Long-Term Sequelae After Acute Pancreatitis.

The clinical utility of body composition in the development of complications of acute pancreatitis (AP) remains unclear. We aimed to describe the associations between body composition and the recurrence of AP. We performed a retrospective study of patients hospitalized with AP at three tertiary care centers. Patients with computer tomography (CT) imaging of the abdomen at admission were included. A previously validated and fully automated abdominal segmentation algorithm was used for body composition analysis. Hospitalization for a recurrent episode of AP was the primary endpoint. Secondary endpoints included the development of chronic pancreatitis (CP) or diabetes mellitus (DM) in patients who were evaluated. Cox Proportional Hazards regression was used. From a total of 347 patients, 89 (25.6%) were hospitalized for recurrent AP (median time: 219days). Thirty-four of 112 patients (30.4%) developed CP (median time: 311days) and 22 of 88 (25.0%) developed DM (median time: 1104days). After adjusting for age, male sex, first episode of AP, BUN, and severity of AP, we found that obesity, body mass index, alcohol pancreatitis, and gallstone pancreatitis were significantly associated with a recurrent episode of AP. Body composition was not associated with recurrent AP. In unadjusted analysis, subcutaneous adipose tissue (SAT) (HR 0.87 per 10 cm2, p = 0.002) was associated with CP. Skeletal muscle (SM) mass approached significance for CP (p = 0.0546). Intermuscular adipose tissue (IMAT) (HR 1.45 per 5 cm2, p = 0.0264) was associated with DM. Body composition was not associated with having a recurrent AP. At follow-up, 30% and 25% of evaluated patients developed CP and DM, respectively. A higher SAT and IMAT were associated with a lower incidence of CP and higher incidence of DM, respectively.

MFP-YOLO: a multi-scale feature perception network for CT bone metastasis detection.

Bone metastasis is one of the most common forms of metastasis in the late stages of malignancy. The early detection of bone metastases can help clinicians develop appropriate treatment plans. CT images are essential for diagnosing and assessing bone metastases in clinical practice. However, early bone metastasis lesions occupy a small part of the image and display variable sizes as the condition progresses, which adds complexity to the detection. To improve diagnostic efficiency, this paper proposes a novel algorithm-MFP-YOLO. Building on the YOLOv5 algorithm, this approach introduces a feature extraction module capable of capturing global information and designs a new content-aware feature pyramid structure to improve the network's capability in processing lesions of varying sizes. Moreover, this paper innovatively applies a transformer-structure decoder to bone metastasis detection. A dataset comprising 3921 CT images was created specifically for this task. The proposed method outperforms the baseline model with a 5.5% increase in precision and a 7.7% boost in recall. The experimental results indicate that this method can meet the needs of bone metastasis detection tasks in real scenarios and provide assistance for medical diagnosis.

YOLO v10-Based Brain Tumor Detection: An Innovative Approach in CT Imaging

YOLO v10-Based Brain Tumor Detection: An Innovative Approach in CT Imaging

The Effect of Systemic Inflammatory Response on Mechanical Thrombectomy is Partly Mediated by Pre-thrombectomy Cerebral Edema in Acute Stroke Patients.

To explore the effect of baseline Systemic Inflammatory Response reflected by platelet-to-lymphocyte ratio (PLR) and pre-thrombectomy cerebral edema reflected by Net Water Uptake (NWU) on futile recanalization in patients with Acute Ischemic Stroke (AIS) after successful thrombectomy, and to investigate the potential mediating role of baseline cerebral edema. 134 Patients with anterior circulation ischemic stroke receiving successful thrombectomy were retrospectively studied. Their demographic and clinical characteristics were collected at admission, and the NWU was quantitatively calculated based on baseline computed tomography (CT). The predictive value of PLR for futile recanalization and the relationship between PLR, NWU, and futile recanalization using mediation analysis were explored. Patients were followed up for 90 days and were divided into a futile recanalization group and a favorable prognosis group [90-day modified Rankin Scale score of 0-2]. High baseline PLR, NWU, no first-pass reperfusion, and large baseline ischemic core volume were independent predictors of futile recanalization after successful thrombectomy in patients with AIS. Mediation analysis results indicate that PLR may partially mediate the occurrence of futile recanalization through NWU. Baseline PLR and NWU were independent predictors of futile recanalization, and higher PLR and NWU values were associated with a higher likelihood of futile recanalization. The findings suggest that early cerebral edema reflected by a high NWU value may be a mediator of PLR-affecting prognosis.

Near Complete Reversal Of Large Diffusion-Weighted Imaging Lesion After Thrombectomy: A Case Report And Literature Review.

Diffusion-Weighted Imaging (DWI) plays a crucial role in acute ischemic stroke (AIS), as it is used to evaluate the ischemic lesions that are irreversibly damaged. The reversibility of DWI ischemic lesions has been noted in patients with AIS who undergo revascularization therapy. In addition, the occurrence of this phenomenon in large ischemic regions remains rare, particularly the near-complete reversal of large DWI lesion cases. A 58-year-old male presented with a generalized tonic-clonic seizure. Emergent magnetic resonance imaging (MRI) revealed an extremely large infarction lesion in the right hemisphere with an Alberta Stroke Program Early Computed Tomographic Score (ASPECTS) value of 2 and occlusion of the terminal right internal carotid artery. The patient was immediately transferred to the Digital Subtraction Angiography (DSA) Unit for endovascular treatment with a stent retriever. After a rapid successful reperfusion with expanded treatment in cerebral infarction (eTICI) score of 3, the patient promptly recovered 24 hours after the procedure. A brain MRI was repeated after 8 days of admission, and interestingly, the DWI lesion showed significant reversal. The modified Rankin scale (mRS) at discharge was 2 and 1 at 90-day follow-up, respectively. Our case shows that the reversibility of DWI ischemic lesions can occur during the acute stroke phase, even in patients with extremely large regions, if rapid and successful reperfusion is achieved. The clinical implications of this phenomenon indicate that using DWI to evaluate the infarct core should be interpreted with caution.

Myofibroblastic Sarcoma of the Maxillary Sinus: Very Rare Entity

Introduction: The Myofibroblastic Sarcoma is a recently discovered malignant tumor that mainly affects the face and neck region, but it occurs most often in the tongue and oral cavity. However, involvement of the maxillary sinus has only been reported a few times in the literature. It is a slowly growing neoplasm with a high recurrence and metastasis risk.  Case Presentation: 18 years old female patient, admitted for a painless mass of the left hemi palate increasing progressively in volume evolving since 5 months, associated to a left nasal obstruction and purulent rhinorrhea. Once in our department, she underwent a facial CT scan, followed by a facial MRI, which showed a tissue process centered primarily on the left maxillary sinus, causing lysis of the inferior wall and bulging of the hard palate on the left side . The surgical decision involved combining an endoscopic and external approach, and was supplemented by radiation therapy. The follow up demonstrated a successful outcome. the patient did not present any recurrence or distant metastasis.  Conclusion: The key lessons are the essential role of imaging in diagnosing and treating LGMS , as well as the crucial role of postoperative radiotherapy in preventing recurrence even though this subject is still controversial.

Ruptured Complex Aortoiliac Aneurysm in an Elderly Patient With a Kidney Transplant Presenting With Sciatica.

Background: Common iliac artery aneurysms are uncommon, with an estimated incidence of less than 0.01% in adults and accounting for only 1% of all intra-abdominal aneurysms. While the risk of rupture is approximately 5%, it increases significantly to 29% once the aneurysm reaches 4 cm. Similarly to abdominal aortic aneurysms, common iliac artery aneurysms often develop silently, remaining asymptomatic in about 70% of cases. This report describes the treatment of a patient with a kidney transplant who underwent endovascular repair of a ruptured left common iliac artery aneurysm with a concomitant abdominal aortic aneurysm and a focal aneurysm of the right renal artery origin. Case Description: A 78-year-old male patient with a kidney transplant presented with left sciatica symptoms and was found to have a contained rupture of a 10 x 7 cm left common iliac artery aneurysm with a concomitant 8 cm abdominal aortic aneurysm and a focal 1.8 cm aneurysm of the right renal artery origin. He underwent an endovascular aneurysm repair with an Aorto-Uni-iliac stent graft and a concomitant right-to-left femoral-femoral bypass using 8 mm ringed PTFE graft and ligation of left external iliac artery to prevent retrograde flow into the left common iliac artery aneurysm. At 3-year follow-up, patient remains stable with a functioning kidney transplant and excluded aneurysms with no evidence of endoleak. Conclusion: This report describes the endovascular repair of a ruptured left common iliac artery aneurysm conducted under local anesthesia in a patient with a renal transplant and complex aneurysm anatomy. The calcification pattern observed on a non-contrast CT scan was effectively used for surgical planning, leading to a successful aneurysm repair while preserving kidney transplant function.

Lightweight Advanced Deep Neural Network (DNN) Model for Early-Stage Lung Cancer Detection

Background: Lung cancer, also known as lung carcinoma, has a high mortality rate; however, an early prediction helps to reduce the risk. In the current literature, various approaches have been developed for the prediction of lung carcinoma (at an early stage), but these still have various issues, such as low accuracy, high noise, low contrast, poor recognition rates, and a high false-positive rate, etc. Thus, in this research effort, we have proposed an advanced algorithm and combined two different types of deep neural networks to make it easier to spot lung melanoma in the early phases. Methods: We have used WDSI (weakly supervised dense instance-level lung segmentation) for laborious pixel-level annotations. In addition, we suggested an SS-CL (deep continuous learning-based deep neural network) that can be applied to the labeled and unlabeled data to improve efficiency. This work intends to evaluate potential lightweight, low-memory deep neural net (DNN) designs for image processing. Results: Our experimental results show that, by combining WDSI and LSO segmentation, we can achieve super-sensitive, specific, and accurate early detection of lung cancer. For experiments, we used the lung nodule (LUNA16) dataset, which consists of the patients’ 3D CT scan images. We confirmed that our proposed model is lightweight because it uses less memory. We have compared them with state-of-the-art models named PSNR and SSIM. The efficiency is 32.8% and 0.97, respectively. The proposed lightweight deep neural network (DNN) model archives a high accuracy of 98.2% and also removes noise more effectively. Conclusions: Our proposed approach has a lot of potential to help medical image analysis to help improve the accuracy of test results, and it may also prove helpful in saving patients’ lives.

Contrast enhancement after mechanical thrombectomy: clinical correlations and impact on outcomes.

Contrast enhancement (CE) after mechanical thrombectomy (MT) remains a subject of investigation, with a reported prevalence ranging from 31 to 88%. We examined our patients to identify predictors of CE and its impact on outcomes, an aspect that remains understudied. We retrospectively analyzed 106 patients who underwent MT at our hospital between 2018 and 2022. The inclusion criteria involved patients who underwent two head CT scans: one immediately after the procedure and a repeat scan within 24h. This study compared demographic, clinical, laboratory, neuroimaging, and procedure-related variables between patients with CE and those without hyperdensity. Regression analysis was employed to determine the associations between CEs and significant variables. Among the analyzed patients, 32.1% had CE. Hyperlipidemia was more prevalent in the CE group. CE correlated with an increase in the National Institutes of Health Stroke Scale (NIHSS) score ≥ 4 within 24h, intracerebral hemorrhage (ICH) incidence, elevated NIHSS score, and a decreased rate of modified Rankin scale (mRS) 0-3 upon discharge. The adjusted model demonstrated a significant association between CE and the incidence of hyperlipidemia and ICH, with an increase in NIHSS score ≥ 4 within 24h of ICH and a lower mRS score of 0-3 upon discharge. CE is associated with hyperlipidemia, ICH, early neurological deterioration, and poor functional outcomes upon discharge. However, no similar association was shown for long-term outcomes. Further studies are required to clarify the pathophysiology of CE and its implications for optimizing stroke care.

Metabolic tumor volume assessed by 18F FDG - PET CT scan as a predictive biomarker for immune checkpoint blockers in advanced NSCLC and its biological correlates.

This study aimed to explore metabolic tumor volume (tMTV) as assessed 18F-fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG-PET/CT), and understand its biological meaning in patients with NSCLC exposed to immune checkpoint blockers(ICBs). In this study, patients with advanced NSCLC and a positive PET scan within 42 days of first line treatment were enrolled in 11 institutions across 4 countries. Total MTV (tMTV) was analyzed, with a 42% SUVmax threshold. Survival was analyzed according to high tMTV (≥ median). Plasma proteomic profile, whole exome, transcriptome and other analysis were performed on monocentric cohorts to explore its biological correlates. Of the 518 patients included, 167 received ICBs, 257 had chemotherapy plus ICBs, and 94 had chemotherapy. Median tMTV was 99 cm3. Median overall survival (OS) for patients with high tMTV treated with ICBs was 11.4 months vs 29.6 months (P<0.0012) for those with low tMTV. In patients receiving chemotherapy-ICB tMTV did not correlate with OS (P=0.099). In patients with PD-L1≥1% and high tMTV, chemotherapy-ICB combination was associated with longer OS compared with ICBs alone (20 vs 11.4 months,p=0.026), while no survival differences observed in low tMTV group. High tMTV correlated (and its detrimental effect seems to be driven by) a specific proteomic profile and increase in genomic instability. Our analysis indicates high tTMV is linked to an increase in systemic inflammation, specific cytokines production and chromosomal instability. tTMV may serve as one of the biomarker to select the best upfront strategy in patients with PD-L1 positive advanced NSCLC.

Locally Advanced Splenic Flexure Tumor With Invasion And Formation Of Splenic Abscess

Introduction: Splenic abscesses are rare, with an incidence of 0.14-0.70% in autopsies, affecting mainly men and immunosuppressed patients. Symptoms include fever, abdominal pain, nausea, and vomiting. Splenic invasion by colon cancer at the splenic flexure is uncommon. We report a case of cancer at the splenic flexure resulting in a splenocolic fistula and splenic abscess. Case Description: A 43-year-old man, with a history of injectable drug use, presented with diffuse abdominal pain. Tests revealed leukocytosis, thrombocytopenia, and elevated bilirubins and liver enzymes. A CT scan identified thickening in the colon and a splenic abscess. Total splenectomy and left colectomy with colon-colonic anastomosis were performed. Discussion: Splenic abscesses are rare and, when associated with colonic neoplasms, present a diagnostic and therapeutic challenge. This case highlights the importance of aggressive surgical intervention to treat tumor invasions and associated complications.

Comparative evaluation of hepatosteatosis in patients with type 2 diabetes mellitus using non-contrast abdominal CT and laboratory findings

Aim: This study aims to underscore the significance of employing multiple parameters from non-contrast abdominal CT scans for the assessment of hepatosteatosis in patients with Type 2 Diabetes Mellitus. Methods: Non-enhanced Computed Tomography of the diabetic subjects were analyzed. Control subjects were selected from non-diabetic patients who had undergone abdominal tomography within the same period. The craniocaudal length of the liver and liver, spleen, pancreas densities, and epicardial adipose tissue were measured. Additionally, patient demographics and laboratory values were retrospectively obtained. Results: The craniocaudal length of the liver was significantly greater in the diabetes mellitus group compared to the control group (168.3 ± 17.2 mm vs 152.3 ± 14.8 mm, p &lt; 0.001). Hepatosteatosis was observed in 22 individuals with diabetes mellitus, whereas only one participant in the control group had this condition (p &lt; 0.001). The diabetes mellitus group exhibited significantly lower median liver density (p &lt; 0.001), liver-spleen density ratio (p = 0.004), pancreatic head density (p = 0.001), and pancreatic body density (p = 0.013). Additionally, the average thickness of epicardial adipose tissue was markedly higher in the diabetes mellitus group compared to the control group (8.1 ± 1.9 mm vs 4.9 ± 1.1 mm, p &lt; 0.001). Conclusions: These data indicate an association between hepatosteatosis and increased epicardial adipose tissue thickness, liver and pancreatic densities in individuals with diabetes mellitus. These findings suggest that non-contrast abdominal CT findings such as epicardial adipose tissue thickness and relevant laboratory tests may aid in evaluating metabolic disorders and fat accumulation in diabetic patients.

Clinical analysis of 20 cases of perinatal tuberculosis

BackgroundTo analyze the clinical manifestations, diagnostic and therapeutic processes of perinatal tuberculosis in children, providing reference for clinicians in the diagnosis and treatment of this disease.MethodsA retrospective analysis was conducted on the epidemiological history, clinical manifestations, laboratory and imaging findings, and treatment follow-up of 20 cases of perinatal tuberculosis diagnosed in the Second Department of Infectious Disease, Kunming Children’s Hospital, from February 2014 to September 2021.ResultsOf the 20 cases, 13 were male (65.0%) and 7 were female (35.0%). The average age at onset was 35.35 ± 23.03days, with an average time from onset to diagnosis of 24.75 ± 15.55days. Tuberculin skin test (TST) was positive in 1 out of 4 cases (25.0%). Gamma interferon release assays (IGRAs) were positive in 9 out of 17 cases (52.9%).acid-fast staining was positive in 7 out of 16 cases (43.7%), and Mycobacterium tuberculosis nucleic acid polymerase-chain-reaction(PCR) was positive in 14 out of 20 cases (70.0%). Chest CT showed miliary changes in 4 out of 19 cases (21.0%), multiple nodular and patchy opacities in 6 out of 19 cases (31.6%), and pulmonary consolidation and atelectasis in 10 out of 19 cases (52.6%). After anti-tuberculosis treatment, 16 out of 20 cases (80.0%) improved, and no recurrence of tuberculosis was observed during follow-up periods ranging from 9 months to 3 years. The growth and development of these children were similar to those of healthy children.ConclusionThe clinical manifestations and chest imaging features of perinatal tuberculosis are nonspecific. In suspected cases, it is crucial to investigate the mother’s condition thoroughly and complete etiological examinations to achieve early diagnosis and timely treatment, which can improve prognosis.

Source free domain adaptation for kidney and tumor image segmentation with wavelet style mining

Kidney cancer is a serious malignant disease, and early diagnosis along with precise segmentation are crucial for effective treatment. However, due to the scarcity of labelled medical image data, the development of the intelligent diagnosis of kidney cancer is restricted. To address this challenge, we propose a novel unsupervised domain adaptation (UDA) framework specifically designed for kidney and tumor CT image segmentation. Our framework consists of a generation phase and an adaptation phase. In the generation phase, we employ a wavelet-based style mining generator to create class-specific source-like images, facilitating domain alignment. In the adaptation phase, we introduce contrastive domain extraction and compact-aware domain consistency modules, enhancing feature-level and output-level adaptability through data augmentation techniques. Experimental results demonstrate that our method performs better in kidney and tumor segmentation tasks, exhibiting higher accuracy and generalization capability than state-of-the-art domain adaptation methods. This indicates that our approach has significant advantages in medical image segmentation for kidneys and tumors.

Noise-assisted hybrid attention networks for low-dose PET and CT denoising.

Positron emission tomography (PET) and computed tomography (CT) play a vital role in tumor-related medical diagnosis, assessment, and treatment planning. However, full-dose PET and CT pose the risk of excessive radiation exposure to patients, whereas low-dose images compromise image quality, impacting subsequent tumor recognition and diseasediagnosis. To solve such problems, we propose a Noise-Assisted Hybrid Attention Network (NAHANet) to reconstruct full-dose PET and CT images from low-dose PET (LDPET) and CT (LDCT) images to reduce patient radiation risks while ensuring the performance of subsequent tumor recognition. NAHANet contains two branches: the noise feature prediction branch (NFPB) and the cascaded reconstruction branch. Among them, NFPB providing noise features for the cascade reconstruction branch. The cascaded reconstruction branch comprises a shallow feature extraction module and a reconstruction module which contains a series of cascaded noise feature fusion blocks (NFFBs). Among these, the NFFB fuses the features extracted from low-dose images with the noise features obtained by NFPB to improve the feature extraction capability. To validate the effectiveness of the NAHANet method, we performed experiments using two public available datasets: the Ultra-low Dose PET Imaging Challenge dataset and Low Dose CT Grand Challengedataset. As a result, the proposed NAHANet achieved higher performance on common indicators. For example, on the CT dataset, the PSNR and SSIM indicators were improved by 4.1dB and 0.06 respectively, and the rMSE indicator was reduced by 5.46 compared with the LDCT; on the PET dataset, the PSNR and SSIM was improved by 3.37dB and 0.02, and the rMSE was reduced by 9.04 compared with theLDPET. This paper proposes a transformer-based denoising algorithm, which utilizes hybrid attention to extract high-level features of low dose images and fuses noise features to optimize the denoising performance of the network, achieving good performance improvements on low-dose CT and PETdatasets.