BACKGROUND: Herpes simplex virus type 2 (HSV-2) is the second most common cause of viral meningitis and the most common cause of recurrent meningitis. Although the incidence of HSV-2 meningitis is high, its clinical characteristics are not well known. The purpose of this study was to review the clinical characteristics and prognosis of HSV-2 meningitis. METHODS: We analyzed patients who were admitted to the Department of Neurology at Severance Hospital with a final diagnosis of HSV-2 meningitis, as confirmed by applying the polymerase chain reaction to the cerebrospinal fluid (CSF) of patients. RESULTS: The study involved 998 patients with aseptic meningitis and 60 patients diagnosed with HSV-2 meningitis. The mean age at meningitis presentation was 32.5 years (range 18-54 years), and 72% of the patients were female. Common clinical symptoms were headache (100%), nausea and/or vomiting (83%), meningismus (57%), and fever (55%). Six patients had a history of genital herpes infection, and 11 had a past history of recurrent meningitis. The CSF study was notable for elevated protein (111.0±53.5 mg/dL, mean±standard deviation) and white cell count (332.0±211.3 cells/µL). The CSF/serum glucose ratio was 0.52±0.90. Various treatments were applied, including conservative care, antiviral agents, empirical antibiotics, and combined treatments. All patients recovered without serious neurologic sequelae. CONCLUSIONS: HSV-2 meningitis is relatively common, as are recurrent episodes. The clinical characteristics of HSV-2 meningitis are similar to those of other types of aseptic meningitis. HSV-2 meningitis is treated using antiviral therapy, and the prognosis is favorable even with conservative treatment.
Anti-Bacterial Agents ; Antiviral Agents ; Cell Count ; Cerebrospinal Fluid ; Diagnosis ; Female ; Fever ; Glucose ; Headache ; Herpes Genitalis ; Herpes Simplex* ; Herpesvirus 2, Human* ; Humans ; Incidence ; Meningism ; Meningitis ; Meningitis, Aseptic ; Meningitis, Viral ; Nausea ; Neurology ; Polymerase Chain Reaction ; Prognosis ; Retrospective Studies* ; Simplexvirus* ; Vomiting

Cranial fasciitis is an uncommon subset of nodular fasciitis composed of spindle cells and myxoid stroma. This is not considered as a true neoplasm, as it occurs mostly in the scalp as a rapidly growing mass accompanied by adjacent bony structure destruction. There are few cases of cranial fasciitis reported in the literature; however, we experienced a case of a 2-year-old girl with swelling of midface. Subtotal resection was performed and the final pathological result confirmed cranial fasciitis. We report this rare case with a review of the literature.
Child, Preschool ; Fasciitis* ; Female ; Humans ; Scalp

Retention cyst of the maxillary sinus is a benign lesion produced from obstruction of a seromucous gland or duct. It is mostly asymptomatic but sometimes is accompanied by facial pain, headache, nasal obstruction, and other symptoms. However, there are some debates on whether the symptoms are directly related with retention cyst. These cysts typically do not require treatment. However, when accompanied by symptoms, treatment can be administered for diagnostic and therapeutic purposes. We report a case in which facial pain is caused by a maxillary retention cyst suspended from an infraorbital nerve.
Facial Pain ; Headache ; Maxillary Sinus ; Nasal Obstruction ; Paranasal Sinus Neoplasms

Neuronal intranuclear inclusion disease (NIID) is a rare neurodegenerative disorder characterized by the presence of eosinophilic nuclear inclusions in neurons and somatic cells. It clinically manifests as cognitive decline, seizures, and autonomic dysfunction. A 44-year-old man presented with a transient visual field defect and hemiparesis. Based on characteristic imaging findings and pathological findings, NIID was suspected and diagnosed through genetic testing. This case emphasizes the importance of comprehensive clinical phenotype analysis and accurate genetic diagnosis.

Neuronal intranuclear inclusion disease (NIID) is a rare neurodegenerative disorder characterized by the presence of eosinophilic nuclear inclusions in neurons and somatic cells. It clinically manifests as cognitive decline, seizures, and autonomic dysfunction. A 44-year-old man presented with a transient visual field defect and hemiparesis. Based on characteristic imaging findings and pathological findings, NIID was suspected and diagnosed through genetic testing. This case emphasizes the importance of comprehensive clinical phenotype analysis and accurate genetic diagnosis.

Neuronal intranuclear inclusion disease (NIID) is a rare neurodegenerative disorder characterized by the presence of eosinophilic nuclear inclusions in neurons and somatic cells. It clinically manifests as cognitive decline, seizures, and autonomic dysfunction. A 44-year-old man presented with a transient visual field defect and hemiparesis. Based on characteristic imaging findings and pathological findings, NIID was suspected and diagnosed through genetic testing. This case emphasizes the importance of comprehensive clinical phenotype analysis and accurate genetic diagnosis.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.