Artificial Intelligence in Pathology.

As in other domains, artificial intelligence is becoming increasingly important in medicine. In particular,deep learning-based pattern recognition methods can advance the field of pathology byincorporating clinical, radiologic, and genomic data to accurately diagnose diseases and predictpatient prognoses. In this review, we present an overview of artificial intelligence, the brief historyof artificial intelligence in the medical domain, recent advances in artificial intelligence applied topathology, and future prospects of pathology driven by artificial intelligence.

Newly Categorized Seromucinous Tumor of the Ovary: Magnetic Resonance Imaging Findings.

OBJECTIVE: The aim of this study was to describe magnetic resonance imaging findings of newly categorized ovarian seromucinous tumors. METHODS: We retrospectively reviewed the images of 29 patients with seromucinous tumor for the following factors: size, configuration, signal intensity (SI), and accompanying ovarian endometriosis. RESULTS: Thirty-two tumors (17 benign, 7 borderline, 8 carcinoma) were found on computed tomography or magnetic resonance imaging. Their mean size was 11.4 cm. Benign tumors appeared as unilocular or multilocular cystic masses. Borderline tumors and carcinomas appeared as complex cystic-solid masses. T2-weighted SI of the solid portion was hyperintense in borderline tumors and intermediate in carcinomas. Endometriosis was present in 18 tumors, and hemorrhage (on images) in 20. CONCLUSIONS: The imaging features of seromucinous tumors varied by tumor type. More than half of tumors were accompanied by endometriosis. High T2-weighted SI of the solid portion could be a specific feature of borderline tumors. It is hard to differentiate seromucinous carcinomas from other endometriosis-related carcinomas.

Paraganglioma in the cauda equina. A case report.

Paraganglioma of the cauda equina is rare, and diagnosis is difficult. A 47-year-old woman presented with backache, with a 2-year history of pain radiating in her right lower extremity. Initial neurologic examination revealed mild hypaesthesia in the L4 dermatome on both sides. Spine MRI showed a well-delinated intradural extramedullary mass compressing the spinal cord. It extended from L2 to L4, with anterior compression of the spinal cord which was displaced posteriorly. Clinical and radiological findings suggested an ependymoma. Surgical decompression was performed from L2 to L4 through lumbar laminectomy under microscope. Intraoperatively, the patient experienced unexplained paroxysmal hypertension while manipulating the tumour, which was not relieved by hypotensive medication but resolved immediately after resection of the mass. Postoperatively, the neurologic status improved and the radiating pain was relieved. Histopathologic examination showed cellular perivascular arrangement which looked like 'pseudorosettes'. Taken together, these histologic and radiologic findings suggested a benign myxopapillary ependymoma. However, immunohistochemical examination showed reactivity with synaptophysin and chromogranin. Finally, histological examination of the specimen revealed a 'Zellballen' pattern of paraganglioma, and the final diagnosis of paraganglioma with secreting function was confirmed. Paraganglioma is a rare tumour that can exhibit a secreting function causing paroxysmal hypertension which may be life threatening. Therefore, the differential diagnosis is important. The diagnosis is based on close examination of the clinical, radiologic and pathologic findings.