Common Imaging Features Research Articles

A Framework for Deep Multitask Learning With Multiparametric Magnetic Resonance Imaging for the Joint Prediction of Histological Characteristics in Breast Cancer.

The clinical management and decision-making process related to breast cancer are based on multiple histological indicators. This study aims to jointly predict the Ki-67 expression level, luminal A subtype and histological grade molecular biomarkers using a new deep multitask learning method with multiparametric magnetic resonance imaging. A multitask learning network structure was proposed by introducing a common-task layer and task-specific layers to learn the high-level features that are common to all tasks and related to a specific task, respectively. A network pretrained with knowledge from the ImageNet dataset was used and fine-tuned with MRI data. Information from multiparametric MR images was fused using the strategy at the feature and decision levels. The area under the receiver operating characteristic curve (AUC) was used to measure model performance. For single-task learning using a single image series, the deep learning model generated AUCs of 0.752, 0.722, and 0.596 for the Ki-67, luminal A and histological grade prediction tasks, respectively. The performance was improved by freezing the first 5 convolutional layers, using 20% shared layers and fusing multiparametric series at the feature level, which achieved AUCs of 0.819, 0.799 and 0.747 for Ki-67, luminal A and histological grade prediction tasks, respectively. Our study showed advantages in jointly predicting correlated clinical biomarkers using a deep multitask learning framework with an appropriate number of fine-tuned convolutional layers by taking full advantage of common and complementary imaging features. Multiparametric image series-based multitask learning could be a promising approach for the multiple clinical indicator-based management of breast cancer.

Covid-19 associated shoulder girdle calcific myositis: a novel entity.

To investigate the prevalence, describe the radiological features, and consider the clinical sequelae of COVID-19- associated shoulder girdle calcific myositis. All patients who underwent a CT pulmonary angiogram study at our institution (Queen Alexandra Hospital, Portsmouth Hospitals University NHS Trust, Portsmouth, United Kingdom) in April and May 2020, January 2021, and July 2021 were included. A total of 1239 CT pulmonary angiogram studies for 1201 patients were reviewed. Patients with COVID-19 and associated shoulder girdle calcific myositis were identified. Their electronic patient records were reviewed. The patients' demographics, serum inflammatory markers, and proning history were recorded. Of the 364 patients in Wave 1, 71 patients (19.5%) had COVID-19, and of those, 2 patients (2.8%) had shoulder girdle calcific myositis. Of the 521 patients in Wave 2, 354 patients (67.9%) had COVID-19, and of those, 3 patients (0.8%) had shoulder girdle calcific myositis. Of the 316 patients in Wave 3, 37 patients (11.7%) had COVID-19, and of those, 1 patient (2.7%) had shoulder girdle calcific myositis. The overall prevalence was 1.3%. The most common site of calcific myositis was within the subscapularis muscle. COVID-19-associated shoulder girdle calcific myositis is a rare extrapulmonary musculoskeletal manifestation of COVID-19. Early recognition and increased awareness of this disease entity, in our experience, aids in reducing patient morbidity and improving long-term functional outcome. We have reported a novel disease entity associated with COVID-19, in the form of shoulder girdle calcific myositis. We have described the common imaging features and discussed our experience of management and clinical sequelae.

Predicting clinical outcome with computed tomography severity score in COVID-19

Introduction: COVID-19 is a major public health burden in the world, and chest computed tomography (CT) is the ideal imaging modality to diagnose and monitor disease progression. Objectives: The objective was to review the common and uncommon chest CT findings of patients with COVID-19 and to correlate the CT findings with short-term prognosis. Methods: All patients who had laboratory-confirmed COVID-19 infection and underwent CT scan were reviewed. The imaging features and the distribution of abnormality were evaluated. A CT severity score (CT-SS) system out of 40 was used. Mann–Whitney U-test and Chi-square or Fisher's exact test were used for analysis. Two-sided P < 0.05 was considered statistically significant. A receiver operating characteristic curve analysis was performed to calculate the CT-SS cutoff for hospital admission and intensive care unit (ICU) admission. Results: A total of 149 individuals were eligible. The most common imaging features were ground-glass opacities (GGO) (88%), septal thickening (70%), and reticulations (50%). The least common imaging features were pneumothorax (1%) and vascular enlargement (1%). The most common distributions of GGO were bilateral (92%) and multifocal (95%), with peripheral (100%) and lower lobe predominance (77%). A higher CT-SS had a significant association with longer hospital stay and ICU admission, with CT-SS of 18 and 10 being optimal cutoff, respectively. Conclusion: Our study is one of the few studies to correlate the imaging finding with clinical outcomes in the south Indian population. The common findings in our study were consistent with the literature. CT-SS plays an important role in predicting prognosis.

Clinical and CT scan findings in patients with COVID-19 pneumonia: a comparison based on disease severity

BackgroundThe SARS-CoV-2 can cause severe pneumonia and highly impact general health. We aimed to investigate different clinical features and CT scan findings of patients with COVID-19 based on disease severity to have a better understanding of this disease.MethodsNinety patients with coronavirus were divided into three categories based on the severity of the disease: mild/moderate, severe, and very severe. Clinical, laboratory, and CT scan findings of the patients were examined retrospectively. Any association between these features and disease severity was assessed.ResultsThe mean age and duration of hospitalization of patients increased with increasing the severity of the disease. The most common clinical symptoms were shortness of breath, cough, and fever. As the severity of the disease increased from mild/moderate to very severe, there was an increase in neutrophil counts and a decrease in lymphocytes and white blood cells (WBC) showing excessive inflammation associated with severe forms of COVID-19. Subpleural changes (81%) and ground-glass opacification/opacity (GGO) lesions (73%) of the lung were the most common features in CT images of COVID-19 patients, and interlobular septal thickening (10%) was the lowest CT feature among patients. Regarding the affected parts of the lung in COVID-19 patients, bilateral, peripheral, and multiple lesions had the highest prevalence.ConclusionsIt has been shown that clinical, laboratory, and CT scan findings varied in COVID-19 patients based on disease severity, which need to be considered carefully in timely diagnosis and treatment of this illness.

Tuberculous myelitis: a prospective follow-up study.

BackgroundProspective studies regarding tuberculous myelitis are lacking. We aimed to prospectively evaluate patients with tuberculous myelitis to identify the features that distinguish tuberculous myelitis from other myelitis.MethodsThis was a prospective study. Patients presenting with paraparesis/quadriparesis, and MRI showing myelitis were included. All patients were subjected to clinical, neuroimaging, and laboratory evaluation. Diagnosis of definite tuberculous myelitis was made if GeneXpert test in CSF was positive. Probable tuberculous myelitis was diagnosed if there was evidence of tuberculosis elsewhere in the body. Patients were treated with methylprednisolone and antituberculosis treatment. Patients were followed for 6 months. We compared the clinical, laboratory, and neuroimaging parameters and response to treatment of tuberculous myelitis with other myelitis. P values were adjusted using the Benjamini-Hochberg (BH) procedure to control false discovery rate.ResultsWe enrolled 52 patients. Eighteen (34.6%) patients had tuberculous myelitis. Headache (P = 0.018) was significantly more common in tuberculous myelitis. The CSF protein (P < 0.001), and CSF cell count (P < 0.001) were significantly higher in tuberculous myelitis. On neuroimaging, a LETM was common in tuberculous myelitis. Spinal meningeal enhancement (14; 77.8%), extra-axial collection, and CSF loculation (6; 33.4%), arachnoiditis (3;16.7%), and concomitant spinal tuberculoma (2;11.1%) were other common imaging features of tuberculous myelitis. Tuberculous myelitis patients showed a better response (P = 0.025) to treatment.ConclusionTuberculous myelitis was seen in approximately 35% of all myelitis cases, in a high tuberculosis endemic zone. Headache, markedly elevated CSF protein and spinal meningeal enhancement were distinguishing features. Tuberculous myelitis patients responded well to corticosteroids.

A New Parameter of Hip Instability in Developmental Dysplasia of the Hip (DDH): Teardrop Distance.

BackgroundHip instability is one of the etiologies of accelerated onset of osteoarthritis in developmental dysplasia of the hip (DDH). There are some radiological parameters for hip instability in hip dysplasia like broken shenton’s line, elevated acetabular index, reduced lateral center edge angle (LCEA), upsloping lateral sourcil. We have discovered a new index of teardrop distance (TD) for assessing instability. Herein, we hypothesized that increased TD could be used as evidence of hip instability in DDH patients, which we verified using TD as an auxiliary diagnostic parameter for DDH, from supine to standing position.MethodsFemale DDH patients undergoing Bernese periacetabular osteotomy (PAO) were enrolled in the DDH group, and normal female volunteers were in the control group. Anteroposterior radiographs of the pelvis in the supine and standing positions were taken, and LCEA, Tönnis angle (TA), sharp angle (SA), and TD were tested using Stata software to analyze the changes between supine and standing anteroposterior pelvic radiographs.ResultsThere were 26 female volunteers with 52 hips in the control group: supine TD 6.80 ± 0.98 mm, standing TD 6.65 ± 1.3 mm (P > 0.05). A total of 78 patients with 135 hips were included in the DDH group: supine TD 10.51 ± 3.50 mm, standing TD 10.93 ± 4.23 mm (P < 0.05). In either supine or standing position, TD in the DDH group was significantly wider than that in the control group (P < 0.05). In the DDH group, TD was correlated with TA and LCEA (rp 0.494–0.588, P < 0.05); TD was not correlated with SA, weight, or BMI (P > 0.05). There was a weak correlation between TD difference and standing LCEA (rp −0.276, P < 0.05).ConclusionTD > 10 mm was a common imaging feature of DDH. It increased from supine to standing position, thus indicating hip instability in DDH patients. The hip parameters of both positions should be compared, fully considering the factors of hip stability.

P.116 Not everything is what it seems, look closer, think deeper: granulomatosis with polyangiitis

Background: Granulomatosis with polyangiitis (GPA) is a rare disease of unknown cause. The multitude of manifestations presents significant diagnostic challenges. This is a unique case of GPA with sinonasal, airway, skull base, petrous bones, vascular, and brain parenchyma involvement. Methods: We present a case of a 45-year-old female with a several day history of headache and left hearing loss. MRI brain demonstrated a large erosive enhancing soft process in the sinonasal cavity and nasopharynx. Results: She developed new ipsilateral rightward tongue deviation. A second MRI demonstrated disease progression. It showed posterior pharyngeal wall ulceration, involvement of the skull base foramina, petrous bones, and central bony skull erosion. It demonstrated right hemiglossal edema secondary to right hypoglossal nerve compression at the skull base. There was enhancing soft tissue partially encasing the left petrous internal carotid artery and more extensively encasing and narrowing bilateral intradural vertebral arteries with associated brainstem edema secondary to direct mass effect and new left occipital thromboembolic infarct. She underwent a nasopharyngeal biopsy which demonstrated acute inflammation on a background of GPA. Conclusions: There are no pathognomonic imaging characteristics for GPA. By recognizing the common and less-common imaging features, radiologists play a crucial role in both diagnosing and monitoring the disease activity.

Clinical and Imaging Features of Tumors in the Scapula.

The scapula is a small irregular-shaped flat bone, which may suffer from a variety of tumors or tumor-like lesions. As the imaging manifestations are complex and changeable, correct imaging diagnosis is difficult. At present, there are few related radiology literatures, and it is necessary to fully analyze the imaging signs of different types of benign and malignant tumors in scapula to guide clinical treatment. This study was to investigate clinical and imaging presentations of tumors and tumor- like lesions in the scapula so as to increase the diagnostic accuracy of diseases in the scapula. Patients with scapular tumors confirmed by pathology were enrolled. The imaging and clinical data were analyzed. Among 108 patients, benign tumors were in 53 (49.1%) cases, intermediate in seven (6.5%), and malignant in 48 (44.4%) involving 16 diseases. Osteochondroma was the first benign tumors in 45 cases accounting for 84.9% of all benign scapular tumors, followed by chondroma in four cases (7.5%). The intermediate tumors were mainly eosinophilic granuloma in four cases. Metastatic tumors were the commonest malignant tumor (27 cases or 56.2% of all malignant tumors), followed by chondrosarcoma (in 13 cases). Except for the one case of chondroblastoma in which the lesion involved the glenoid cavity, all the other cartilaginous tumors were located in the scapular body and processes. The type of lesions in the bony processes is the same as in the scapular body, the common lesions in the central area of the body were malignant tumors, and the commonest lesions in the glenoid area were metastasis. Common imaging features of malignant scapular tumors were ill-defined margins, cortical destruction and soft tissue involvement. The imaging features of chondrosarcoma lack specificity except for calcification. Benign lesions usually had a clear boundary and marginal sclerosis. A wide variety of benign and malignant tumors may occur in the scapula with mostly cartilaginous and metastatic tumors, and the location and distribution of lesions are similar in the scapula to those in the long bones.

Improving Multi-View Camera Calibration Using Precise Location of Sphere Center Projection

Several calibration algorithms use spheres as calibration tokens because of the simplicity and uniform shape that a sphere presents across multiple views, along with the simplicity of its construction. Among the alternatives are the use of complex 3D tokens with reference marks, usually complex to build and analyze with the required accuracy; or the search of common features in scene images, with this task being of high complexity too due to perspective changes. Some of the algorithms using spheres rely on the estimation of the sphere center projection obtained from the camera images to proceed. The computation of these projection points from the sphere silhouette on the images is not straightforward because it does not match exactly the silhouette centroid. Thus, several methods have been developed to cope with this calculation. In this work, a simple and fast numerical method adapted to precisely compute the sphere center projection for these algorithms is presented. The benefits over other similar existing methods are its ease of implementation and that it presents less sensibility to segmentation issues. Other possible applications of the proposed method are presented too.

CT FINDINGS OF MALIGNANT NEOPLASTIC MASSES OF SMALL AND LARGE INTESTINE WITH HISTOPATHOLOGICAL CORRELATION: A CROSS SECTIONAL STUDY

AIMS AND OBJECTIVES: To formulate an appropriate scanning protocol technique to evaluate malignant neoplasm of small and large intestine. To dene various CT ndings in malignant bowel neoplasm. To stage bowel neoplasm. To accurately characterize the tumour by histologic type as far as possible. Materials And Methods: In our cross-sectional study, 76 cases of small and large bowel carcinoma were referred to radiology department. After taking informed consent, CECT abdomen of the patients was done. All these patients went for surgical excision of the neoplastic mass; and the mass were sent for HPE examination. Epidemiological data, clinical features, CECT abdominal features were correlated with HPE type of the mass were found out. P-Value&lt;0.05 was considered to be statistically signicant. Results: In Small bowel, ileum was the commonest site of involvement (5.26%), whereas sigmoid colon and rectum (18.42%) was commonest site of involvement in large bowel. Adenocarcinoma was the commonest histological subtype seen in 54 patients (71.05%), followed by lymphoma in 14 patients (18.42%), Mucinous Adenocarcinoma seen in 5 patients (57.5%). and gastrointestinal stromal tumor [GIST] seen in 3 patients(3.94%). Conclusion: Common CT imaging features in malignant bowel neoplasm are bowel Wall thickening &gt;2cm,short Segment Involvementand heterogeneous mixed pattern. Although CT has a high specicity for detection of a metastatic lymph nodes, sensitivity is low.Liver is the commonest site of distant metastasis in malignant bowel neoplasm. Helical CT is a noninvasive, reliable, and accurate technique for imaging the liver and should be considered as the standard preoperative work-up of hepatic metastases from bowel malignant.

Imaging of the head and neck during the COVID19 pandemic

There is a wide spectrum of clinical manifestation of COVID-19 in the head and neck, but often these do not have an imaging correlate. This review will highlight the most common imaging features of COVID-19 in the head and neck that can be seen on routine head and neck CT and MRI. In addition, situations where a more dedicated imaging protocol is required will be highlighted. Finally, as mass vaccination efforts are underway worldwide, post vaccination imaging can often complicate cancer surveillance imaging. Post vaccination imaging features and recommendations will be discussed.

Imaging of Abusive Head Trauma : A Radiologists' Perspective.

Abusive head trauma (AHT) is the most common and serious form of child abuse and a leading cause of traumatic death in infants and young children. The biomechanics of head injuries include violent shaking, blunt impact, or a combination of both. Neuroimaging plays an important role in recognizing and distinguishing abusive injuries from lesions from accidental trauma or other causes, because clinical presentation and medical history are often nonspecific and ambiguous in this age group. Understanding common imaging features of AHT can increase recognition with high specificity for AHT. In this review, we discuss the biomechanics of AHT, imaging features of AHT, and other conditions that mimic AHT.

Imaging features in extramedullary hematopoiesis - A case series

Extramedullary hematopoiesis (EMH) is a rare compensatory process associated with many hematologic disorders and bone marrow dysfunction. Through this article, we aim to present various imaging features of EMH and its complications including spinal cord compression. Three cases of EMH in hemoglobinopathies (thalassemia and sickle cell anemia) have been described. Cases 1 and 2 of our series include paraspinal masses causing cord compression. Case 1 of thalassemia intermedia presented with various skeletal deformities in long bones, skull, and facial bones (Chipmunk facies). Case 2 was diagnosed with extrapleural lung mass with other paraspinal lesions. Case 3 presented with multiple enlarged retroperitoneal lymph nodes due to EMH. Common and uncommon imaging features of EMH have been described. Lymph node involvement was difficult to diagnose as it mimicked tubercular and neoplastic etiologies; hence, histopathological diagnosis was mandatory.

Use of a DNN-Based Image Translator with Edge Enhancement Technique to Estimate Correspondence between SAR and Optical Images

In this paper, the local correspondence between synthetic aperture radar (SAR) images and optical images is proposed using an image feature-based keypoint-matching algorithm. To achieve accurate matching, common image features were obtained at the corresponding locations. Since the appearance of SAR and optical images is different, it was difficult to find similar features to account for geometric corrections. In this work, an image translator, which was built with a DNN (deep neural network) and trained by conditional generative adversarial networks (cGANs) with edge enhancement, was employed to find the corresponding locations between SAR and optical images. When using conventional cGANs, many blurs appear in the translated images and they degrade keypoint-matching accuracy. Therefore, a novel method applying an edge enhancement filter in the cGANs structure was proposed to find the corresponding points between SAR and optical images to accurately register images from different sensors. The results suggested that the proposed method could accurately estimate the corresponding points between SAR and optical images.

Imaging of Cancer Immunotherapy: Response Assessment Methods, Atypical Response Patterns, and Immune-Related Adverse Events, From the AJR Special Series on Imaging of Inflammation.

The introduction of immunotherapy with immune-checkpoint inhibitors (ICIs) has revolutionized cancer treatment paradigms. Since FDA approval of the first ICI in 2011, multiple additional ICIs have been approved and granted marketing authorization, and many promising agents are in early clinical adoption. Due to the distinctive biologic mechanisms of ICIs, the patterns of tumor response and progression seen with immunotherapy differ from those observed with cytotoxic chemothera-pies. With increasing clinical adoption of immunotherapy, it is critical for radiologists to recognize different response patterns and common pitfalls to avoid misinterpretation of imaging studies or prompt premature cessation of potentially effective treatment. This review provides an overview of ICIs and their mechanisms of action and discusses anatomic and metabolic immune-related response assessment methods, typical and atypical patterns of immunotherapy response (including pseudoprogression, hyperprogression, dissociated response, and durable response), and common imaging features of immune-related adverse events. Future multicenter trials are needed to validate the proposed immune-related response criteria and identify the functional imaging markers of early treatment response and survival.

The Integral Role of Radiology in the Diagnosis and Management of COVID-19-Associated Mucormycosis Infections.

ABSTRACT.Recently, there has been a sudden surge in COVID-19–associated mucormycosis (CAM) infections. Rhino-oculo-cerebral and pulmonary mucormycosis are the two most common forms of CAM. Radiology plays an integral role in the management of CAM. Computed tomography (CT) determines gross bony and soft tissue involvement in COVID-19–associated rhino-oculo-cerebral mucormycosis, whereas magnetic resonance imaging helps in evaluation of the orbital and intracranial extension. Paranasal sinus soft tissue with extrasinus infiltration with or without bony destruction is suggestive of COVID-19–associated rhino-oculo-cerebral mucormycosis. High-resolution CT chest scan has shown to be helpful in the diagnosis of COVID-19–associated pulmonary mucormycosis. Consolidation and cavitation are the most common imaging features. Other CT abnormalities include the reverse-halo sign, pleural effusion, ground-glass opacities, pneumothorax, nodules, and pulmonary embolism. A high index of suspicion with appropriate imaging findings can lead to the early diagnosis of CAM and timely initiation of antifungal treatment and/or surgical debridement, which can be lifesaving.

Clinical and Imaging Characteristics of Primary Severe Community-Acquired Pneumonia Caused by Hypervirulent Klebsiella Pneumoniae.

To investigate the CT imaging features and microbial phenotypes of primary severe community-acquired pneumonia caused by hypervirulent Klebsiella pneumoniae (hvKp). Patients diagnosed with primary hvKp pneumonia were included, and their clinical data were analyzed, including the baseline characteristics and CT imaging results. After hypermucoviscosity phenotyping, the strains, serological types, and virulence genes of hvKp were identified using multiplex PCR. Twelve patients with primary hvKp pneumonia were included (11 males, 1 female). All patients were infected via respiratory tract inhalation. Ten patients were long-term drinkers. Four patients (33.3%), who were long-term alcohol abusers, died within 30 days after diagnosis. No extrapulmonary metastatic infection was found in any patient. The imaging of lung lesions at the early disease stage exhibited an extensive consolidation in the lungs. As the disease progressed, the most common imaging features were pleural effusion (9/12), cavitation and necrosis (8/12), and pneumothorax (3/12). The serological typing of the capsular polysaccharides on hvKp strains were K1 (6/12) and K2 (6/12). Furthermore, the virulence genotyping showed rmpA (11/12), magA (11/12), ureA (12/12), mrkD (12/12), fim-1 (12/12), wabG (12/12), ybtS (12/12), and iucB (11/12). Primary severe community-acquired hvKp-associated pneumonia is more common in men, especially those with a long-term history of alcohol consumption. CT scanning at the early disease stage mostly showed extensive pulmonary consolidation, which was prone to be combined with cavitation, necrosis, and pleural effusion. K1 and K2 serotypes were identified among the hvKp strains, which were not prone to form extrapulmonary metastasis via the bloodstream.

An Automatic Method for Core Orientation Based on Planar Geologic Features in Drill-Core Scans and Microresistivity Images

Drilling cores, the most direct and accurate first-hand data reflecting the subsurface geological information, are of unclear orientation for most boreholes. The image feature comparison method is one of the most widely used methods for core orientation. However, this method is mainly conducted by manual comparison, which is labor-intensive, time-consuming and inaccurate. In this paper, we propose an automatic core orientation method by comparing the planar geologic features (one of the most common features) in drill-core scans and microresistivity images. This method first automatically detects planar geologic features by combing the vertical gradient-based Otsu threshold segmentation, Hough transform and optimal sinusoid extraction and then determines correction difference of the core orientation by comparing the dip azimuths of the optimal sinusoids from the two types of images. This method is successfully applied to actual data. The automatic core orientation results are compared with the manual core orientation results. The results show that this method can eliminate human errors and has higher accuracy, especially for core scan images with non-distinct features. The proposed method has wide application potential in the fields of hydrocarbon exploration and development, scientific drilling research, and basic geological surveys.

Research progress of bone marrow edema-like lesions in knee osteoarthritis

Knee osteoarthritis-associated bone marrow edema-like lesions (KOA-BMLs) is a common MRI imaging feature, which is mainly manifested as abnormal bone marrow hyperintensity in subchondral bone on T2 imaging. The formation of KOA-BMLs may be related to the abnormality of lower limb force line and subchondral bone perfusion, and related histopathological studies showed that the remodeling of bone and bone marrow in these damaged areas was abnormally increased. In KOA patients, the size of BMLs can fluctuate or even disappear in a relatively short period of time, and was closely related to pain, subchondral bone cyst formation, and the progression of KOA. However, the current treatment methods for KOA-BMLs are limited, and there is no uniform guideline or expert consensus, mainly includingmedication, physical therapy and surgical treatment. This article reviews the research progress of the disease characteristics and treatment of KOA-BMLs in order to provide guidance for the clinical diagnosis and treatment of KOA-BMLs.

DenseSENet: more accurate and robust cross-domain iris recognition

With the broad application of iris recognition systems, iris images acquired under cross-sensor and cross-spectrum conditions are increasingly common. Differences in the information stored in these images place high requirements for accuracy and robustness on cross-domain iris recognition algorithms, and current algorithms have not performed well in cross-domain iris recognition. Hence, we propose a dense squeeze and excitation network (DenseSENet) to extract common features in cross-domain iris images and use real-valued features for cross-domain matching. DenseSENet is a deep convolutional neural network based on an attention mechanism. It uses the batch normalization-rectified linear unit-Conv1 × 1 structure to increase the nonlinearity of the network and repeatedly uses features through dense connections to effectively prevent overfitting. The proposed dense squeeze and excitation structure extracts more accurate and stable features by learning the relations between channels and assigning corresponding attention values. The structure of global average pooling-fully connected-softmax is utilized to obtain compact and normalized real-valued feature templates to effectively reduce the size of the iris template. The proposed cross-domain iris recognition framework based on DenseSENet is end-to-end, which transforms the cross-domain matching problem to a many-to-one classification problem. Experiments are conducted on publicly available cross-sensor and cross-spectral iris databases. The results show that the proposed real-valued feature matching based on DenseSENet has better matching accuracy and robustness than the state-of-art cross-domain matching algorithm on two publicly available cross-sensor and cross-spectral iris databases.