-
IEEE Transactions on Medical Imaging Jul 2024A large-scale labeled dataset is a key factor for the success of supervised deep learning in most ophthalmic image analysis scenarios. However, limited annotated data is...
A large-scale labeled dataset is a key factor for the success of supervised deep learning in most ophthalmic image analysis scenarios. However, limited annotated data is very common in ophthalmic image analysis, since manual annotation is time-consuming and labor-intensive. Self-supervised learning (SSL) methods bring huge opportunities for better utilizing unlabeled data, as they do not require massive annotations. To utilize as many unlabeled ophthalmic images as possible, it is necessary to break the dimension barrier, simultaneously making use of both 2D and 3D images as well as alleviating the issue of catastrophic forgetting. In this paper, we propose a universal self-supervised Transformer framework named Uni4Eye++ to discover the intrinsic image characteristic and capture domain-specific feature embedding in ophthalmic images. Uni4Eye++ can serve as a global feature extractor, which builds its basis on a Masked Image Modeling task with a Vision Transformer architecture. On the basis of our previous work Uni4Eye, we further employ an image entropy guided masking strategy to reconstruct more-informative patches and a dynamic head generator module to alleviate modality confusion. We evaluate the performance of our pre-trained Uni4Eye++ encoder by fine-tuning it on multiple downstream ophthalmic image classification and segmentation tasks. The superiority of Uni4Eye++ is successfully established through comparisons to other state-of-the-art SSL pre-training methods. Our code is available at Github.
PubMed: 38954581
DOI: 10.1109/TMI.2024.3422102 -
Epileptic Disorders : International... Jul 2024Mild malformation of cortical development with oligodendroglial hyperplasia and epilepsy (MOGHE) is a recently described, histopathologically and molecularly defined...
OBJECTIVE
Mild malformation of cortical development with oligodendroglial hyperplasia and epilepsy (MOGHE) is a recently described, histopathologically and molecularly defined (SLC35A2-mutated) type of cortical malformation. Although increasingly recognized, the diagnosis of MOGHE remains a challenge. We present the characteristics of the first six patients diagnosed in Bulgaria, with the aim to facilitate identification, proper presurgical evaluation, and surgical treatment approach in this disease.
METHODS
Revision of histopathological specimens of 202 patients operated on for drug-resistant focal epilepsy identified four cases with MOGHE. Another two were suggested, based on clinical characteristics and subsequently, were histologically confirmed. Sanger SLC35A2 sequencing on paraffin-embedded or fresh-frozen brain tissue was performed. Analysis of seizure types, neuropsychological profiles, electroencephalographic (EEG), imaging features and epilepsy surgery outcomes was done.
RESULTS
Three out of the six cases (50%) harbored pathogenic SLC35A2 mutations. One patient had a heterozygous somatic variant with uncertain significance. Clinical characteristics included epilepsy onset in infancy (in 100% under 3 years of age), multiple seizure types, and moderate or severe intellectual/developmental delay. Epileptic spasms with hypsarrhythmia on EEG were the initial seizure type in five patients. The subsequent seizure types resembled those in Lennox-Gastaut syndrome. The majority of the patients (n = 4) presented prominent and persisting autistic features. Magnetic resonance imaging (MRI) showed multilobar (n = 6) and bilateral (n = 3) lesions, affecting the frontal lobes (n = 5; bilaterally in three) and characterized by increased signal on T2/fluid-attenuated inversion recovery (FLAIR). Voxel-based morphometric MRI post-processing and positron emission tomography helped determining the localization and extent of the lesions and presumed epileptogenic zones. After surgery, four patients (66.7%) were seizure-free ≥2 years. Interestingly, all seizure-free patients carried somatic SLC35A2-alterations.
SIGNIFICANCE
Epileptic spasms, early prominent neuropsychological disturbances, MRI-T2/FLAIR hyperintense lesions with cortico-subcortical blurring, frequently multilobar and especially frontal, can preoperatively help to suspect MOGHE. Epilepsy surgery is still the only successful treatment option in MOGHE.
PubMed: 38953904
DOI: 10.1002/epd2.20261 -
The Canadian Veterinary Journal = La... Jul 2024A 7-year-old spayed female domestic shorthair cat was presented for evaluation of a large-volume abdominal space-occupying lesion. A computed tomography angiography...
A 7-year-old spayed female domestic shorthair cat was presented for evaluation of a large-volume abdominal space-occupying lesion. A computed tomography angiography examination detected a round retroperitoneal mass, in contact with the large abdominal vessels, characterized by an external hyperattenuating capsule and a larger hypoattenuating center. The capsule was soft-tissue attenuating with marked heterogenous contrast enhancement. The center was hypoattenuating pre- and post-contrast administration. The mass displaced both kidneys laterally and the descendent colon ventrally. The mesenteric veins and both phrenicoabdominal veins were markedly increased in diameter. However, the adrenals were not involved. On the excretory phase, no contrast enhancement was observed in either ureter, except for the proximal tract of the right ureter. At laparotomy, both ureters entered the mass that was adherent to the great abdominal vessels. The cytological diagnosis was retroperitoneal extra-adrenal paraganglioma. In cats, retroperitoneal extra-adrenal paragangliomas are very rare. This is the first computed tomography angiography report of a retroperitoneal extra-adrenal paraganglioma in a domestic cat. Key clinical message: This report describes the computed tomography angiography features of a rare case of a retroperitoneal extra-adrenal paraganglioma in a cat. These features could be taken into consideration to direct the diagnosis of a possible neuroendocrine origin for a retroperitoneal mass in a cat.
Topics: Animals; Cats; Female; Cat Diseases; Retroperitoneal Neoplasms; Paraganglioma, Extra-Adrenal; Computed Tomography Angiography
PubMed: 38952751
DOI: No ID Found -
IScience Jun 2024To study neurovascular function in type 2 diabetes mellitus (T2DM), we established a high-fat diet/streptozotocin (HFD/STZ) rat model. Electrocorticography-laser speckle...
To study neurovascular function in type 2 diabetes mellitus (T2DM), we established a high-fat diet/streptozotocin (HFD/STZ) rat model. Electrocorticography-laser speckle contrast imaging (ECoG-LSCI) revealed that the somatosensory-evoked potential (SSEP) amplitude and blood perfusion volume were significantly lower in the HFD/STZ group. Cortical spreading depression (CSD) velocity was used as a measure of neurovascular function, and the results showed that the blood flow velocity and the number of CSD events were significantly lower in the HFD/STZ group. In addition, to compare changes during acute hyperglycemia and hyperglycemia, we used intraperitoneal injection (IPI) of glucose to induce transient hyperglycemia. The results showed that CSD velocity and blood flow were significantly reduced in the IPI group. The significant neurovascular changes observed in the brains of rats in the HFD/STZ group suggest that changes in neuronal apoptosis may play a role in altered glucose homeostasis in T2DM.
PubMed: 38952685
DOI: 10.1016/j.isci.2024.110108 -
International Journal of Nanomedicine 2024How to ingeniously design multi-effect photosensitizers (PSs), including multimodal imaging and multi-channel therapy, is of great significance for highly spatiotemporal...
BACKGROUND
How to ingeniously design multi-effect photosensitizers (PSs), including multimodal imaging and multi-channel therapy, is of great significance for highly spatiotemporal controllable precise phototherapy of malignant tumors.
METHODS
Herein, a novel multifunctional zinc(II) phthalocyanine-based planar micromolecule amphiphile () was successfully designed and synthesized, in which N atom with photoinduced electron transfer effect was introduced to enhance the near-infrared absorbance and nonradiative heat generation. After simple self-assembling into nanoparticles (NPs), would exhibit enhanced multimodal imaging properties including fluorescence (FL) imaging (FLI) /photoacoustic (PA) imaging (PAI) /infrared (IR) thermal imaging, which was further used to guide the combined photodynamic therapy (PDT) and photothermal therapy (PTT).
RESULTS
It was that under the self-guidance of the multimodal imaging, could precisely pinpoint the tumor from the vertical and horizontal boundaries achieving highly efficient and accurate treatment of cancer.
CONCLUSION
Accordingly, the integration of FL/PA/IR multimodal imaging and PDT/PTT synergistic therapy pathway into one could provide a blueprint for the next generation of phototherapy, which offered a new paradigm for the integration of diagnosis and treatment in tumor and a promising prospect for precise cancer therapy.
Topics: Photosensitizing Agents; Multimodal Imaging; Isoindoles; Animals; Humans; Indoles; Photochemotherapy; Nanoparticles; Mice; Zinc Compounds; Organometallic Compounds; Cell Line, Tumor; Photoacoustic Techniques; Photothermal Therapy; Neoplasms; Mice, Inbred BALB C; Phototherapy; Female
PubMed: 38952677
DOI: 10.2147/IJN.S461843 -
Molecular Imaging 2024Chimeric antigen receptor (CAR)-T cell-based immunotherapy has emerged as a path-breaking strategy for certain hematological malignancies. Assessment of the response to... (Review)
Review
Chimeric antigen receptor (CAR)-T cell-based immunotherapy has emerged as a path-breaking strategy for certain hematological malignancies. Assessment of the response to CAR-T therapy using quantitative imaging techniques such as positron emission tomography/computed tomography (PET/CT) has been broadly investigated. However, the definitive role of PET/CT in CAR-T therapy remains to be established. [F]FDG PET/CT has demonstrated high sensitivity and specificity for differentiating patients with a partial and complete response after CAR-T therapy in lymphoma. The early therapeutic response and immune-related adverse effects such as cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome can also be detected on [F]FDG PET images. In otherwise asymptomatic lymphoma patients with partial response following CAR-T therapy, the only positive findings could be abnormal PET/CT results. In multiple myeloma, a negative [F]FDG PET/CT after receiving B-cell maturation antigen-directed CAR-T therapy has been associated with a favorable prognosis. In leukemia, [F]FDG PET/CT can detect extramedullary metastases and treatment responses after therapy. Hence, PET/CT is a valuable imaging tool for patients undergoing CAR-T therapy for pretreatment evaluation, monitoring treatment response, assessing safety, and guiding therapeutic strategies. Developing guidelines with standardized cutoff values for various PET parameters and tumor cell-specific tracers may improve the efficacy and safety of CAR-T therapy.
Topics: Humans; Positron Emission Tomography Computed Tomography; Hematologic Neoplasms; Immunotherapy, Adoptive; Immunotherapy; Receptors, Chimeric Antigen; Fluorodeoxyglucose F18
PubMed: 38952399
DOI: 10.1177/15353508241257924 -
Molecular Imaging 2024This meeting report summarizes a consultants meeting that was held at International Atomic Energy Agency Headquarters, Vienna, in July 2022 to provide an update on the...
This meeting report summarizes a consultants meeting that was held at International Atomic Energy Agency Headquarters, Vienna, in July 2022 to provide an update on the development of multimodality imaging by combining nuclear medicine imaging agents with other nonradioactive molecular probes and/or biomedical imaging techniques.
Topics: Nuclear Medicine; Multimodal Imaging; Humans
PubMed: 38952398
DOI: 10.1177/15353508241245265 -
Zhongguo Xue Xi Chong Bing Fang Zhi Za... Jun 2024To investigate the feasibility of developing a grading diagnostic model for schistosomiasis-induced liver fibrosis based on B-mode ultrasonographic images and clinical...
OBJECTIVE
To investigate the feasibility of developing a grading diagnostic model for schistosomiasis-induced liver fibrosis based on B-mode ultrasonographic images and clinical laboratory indicators.
METHODS
Ultrasound images and clinical laboratory testing data were captured from schistosomiasis patients admitted to the Second People's Hospital of Duchang County, Jiangxi Province from 2018 to 2022. Patients with grade I schistosomiasis-induced liver fibrosis were enrolled in Group 1, and patients with grade II and III schistosomiasis-induced liver fibrosis were enrolled in Group 2. The machine learning binary classification tasks were created based on patients'radiomics and clinical laboratory data from 2018 to 2021 as the training set, and patients'radiomics and clinical laboratory data in 2022 as the validation set. The features of ultrasonographic images were labeled with the ITK-SNAP software, and the features of ultrasonographic images were extracted using the Python 3.7 package and PyRadiomics toolkit. The difference in the features of ultrasonographic images was compared between groups with test or Mann-Whitney test, and the key imaging features were selected with the least absolute shrinkage and selection operator (LASSO) regression algorithm. Four machine learning models were created using the Scikit-learn repository, including the support vector machine (SVM), random forest (RF), linear regression (LR) and extreme gradient boosting (XGBoost). The optimal machine learning model was screened with the receiver operating characteristic curve (ROC), and features with the greatest contributions to the differentiation features of ultrasound images in machine learning models with the SHapley Additive exPlanations (SHAP) method.
RESULTS
The ultrasonographic imaging data and clinical laboratory testing data from 491 schistosomiasis patients from 2019 to 2022 were included in the study, and a total of 851 radiomics features and 54 clinical laboratory indicators were captured. Following statistical tests ( = -5.98 to 4.80, = 6 550 to 20 994, all values < 0.05) and screening of key features with LASSO regression, 44 features or indicators were included for the subsequent modeling. The areas under ROC curve (AUCs) were 0.763 and 0.611 for the training and validation sets of the SVM model based on clinical laboratory indicators, 0.951 and 0.892 for the training and validation sets of the SVM model based on radiomics, and 0.960 and 0.913 for the training and validation sets of the multimodal SVM model. The 10 greatest contributing features or indicators in machine learning models included 2 clinical laboratory indicators and 8 radiomics features.
CONCLUSIONS
The multimodal machine learning models created based on ultrasound-based radiomics and clinical laboratory indicators are feasible for intelligent identification of schistosomiasis-induced liver fibrosis, and are effective to improve the classification effect of one-class data models.
Topics: Humans; Schistosomiasis; Liver Cirrhosis; Machine Learning; Ultrasonography; Male; Female; Middle Aged; Adult; Support Vector Machine; Image Processing, Computer-Assisted; Radiomics
PubMed: 38952311
DOI: 10.16250/j.32.1374.2024110 -
Advances in Computational Biology for Diagnosing Neurodegenerative Diseases: A Comprehensive Review.Zhongguo Ying Yong Sheng Li Xue Za Zhi... Jul 2024The numerous and varied forms of neurodegenerative illnesses provide a considerable challenge to contemporary healthcare. The emergence of artificial intelligence has... (Review)
Review
The numerous and varied forms of neurodegenerative illnesses provide a considerable challenge to contemporary healthcare. The emergence of artificial intelligence has fundamentally changed the diagnostic picture by providing effective and early means of identifying these crippling illnesses. As a subset of computational intelligence, machine-learning algorithms have become very effective tools for the analysis of large datasets that include genetic, imaging, and clinical data. Moreover, multi-modal data integration, which includes information from brain imaging (MRI, PET scans), genetic profiles, and clinical evaluations, is made easier by computational intelligence. A thorough knowledge of the course of the illness is made possible by this consolidative method, which also facilitates the creation of predictive models for early medical evaluation and outcome prediction. Furthermore, there has been a great deal of promise shown by the use of artificial intelligence to neuroimaging analysis. Sophisticated image processing methods combined with machine learning algorithms make it possible to identify functional and structural anomalies in the brain, which often act as early indicators of neurodegenerative diseases. This chapter examines how computational intelligence plays a critical role in improving the diagnosis of neurodegenerative diseases such as Parkinson's, Alzheimer's, etc. To sum up, computational intelligence provides a revolutionary approach for improving the identification of neurodegenerative illnesses. In the battle against these difficult disorders, embracing and improving these computational techniques will surely pave the path for more individualized therapy and more therapies that are successful.
Topics: Humans; Neurodegenerative Diseases; Computational Biology; Neuroimaging; Machine Learning; Algorithms; Artificial Intelligence; Brain; Image Processing, Computer-Assisted; Magnetic Resonance Imaging
PubMed: 38952174
DOI: 10.62958/j.cjap.2024.008 -
BMC Cardiovascular Disorders Jul 2024Pulmonary embolisms (PEs) exhibit clinical features similar to those of acute coronary syndrome (ACS), including electrocardiographic abnormalities and elevated troponin... (Review)
Review
BACKGROUND
Pulmonary embolisms (PEs) exhibit clinical features similar to those of acute coronary syndrome (ACS), including electrocardiographic abnormalities and elevated troponin levels, which frequently lead to misdiagnoses in emergency situations.
CASE PRESENTATION
Here, we report a case of PE coinciding with chronic coronary syndrome in which the patient's condition was obscured by symptoms mimicking ACS. A 68-year-old female with syncope presented to the hospital. Upon admission, she was found to have elevated troponin levels and an electrocardiogram showing ST-segment changes across multiple leads, which initially led to a diagnosis of ACS. Emergency coronary arteriography revealed occlusion of the posterior branches of the left ventricle of the right coronary artery, but based on the complexity of the intervention, the occlusion was considered chronic rather than acute. On the 3rd day after admission, the patient experienced recurrent chest tightness and shortness of breath, which was confirmed as acute PE by emergency computed tomography pulmonary angiography. Following standardized anticoagulation treatment, the patient improved and was subsequently discharged.
CONCLUSIONS
This case report highlights the importance of recognizing the nonspecific features of PE. Clinicians should be vigilant when identifying other clinical features that are difficult to explain accompanying the expected disease, and it is necessary to carefully identify the causes to prevent missed diagnoses or misdiagnoses.
Topics: Humans; Pulmonary Embolism; Female; Aged; Acute Coronary Syndrome; Diagnosis, Differential; Predictive Value of Tests; Electrocardiography; Computed Tomography Angiography; Anticoagulants; Coronary Angiography; Chronic Disease; Treatment Outcome; Diagnostic Errors; Biomarkers
PubMed: 38951773
DOI: 10.1186/s12872-024-03998-6