記述言語：英語   掲載種別：研究論文（学術雑誌）  

BACKGROUND: Although previous studies have demonstrated that tumor deposits (TDs) are associated with worse prognosis in colon cancer, their clinical significance in rectal cancer has not been fully elucidated, especially in the lateral pelvic lymph node (LPLN) area. This study aimed to clarify the clinical significance of TDs, focusing on the number of metastatic foci, including lymph node metastases (LNMs) and TDs, in the LPLN area. METHODS: This retrospective study involved 226 consecutive patients with cStage II/III low rectal cancer who underwent LPLN dissection. Metastatic foci, including LNM and TD, in the LPLN area were defined as lateral pelvic metastases (LP-M) and were evaluated according to LP-M status: presence (absence vs. presence), histopathological classification (LNM vs. TD), and number (one to three vs. four or more). We evaluated the relapse-free survival of each model and compared them using the Akaike information criterion (AIC) and Harrell's concordance index (c-index). RESULTS: Forty-nine of 226 patients (22%) had LP-M, and 15 patients (7%) had TDs. The median number of LP-M per patient was one (range, 1-9). The best risk stratification power was observed for number (AIC, 758; c-index, 0.668) compared with presence (AIC, 759; c-index, 0.665) and histopathological classification (AIC, 761; c-index, 0.664). The number of LP-M was an independent prognostic factor for both relapse-free and overall survival, and was significantly associated with cumulative local recurrence. CONCLUSION: The number of metastatic foci, including LNMs and TDs, in the LPLN area is useful for risk stratification of patients with low rectal cancer.

DOI： 10.1007/s10147-023-02391-1

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

The examination of urinary sediment crystals, the sedimentary components of urine, is useful in screening tests, and is always performed in medical examinations. The examination of urinary sediment crystals is typically done by classifying them under a microscope. Although automated analyzers are commercially available, manual classification is required, which is time-consuming and varies depending on the technologist performing the test and the laboratory. A set of test images was created, consisting of training, validation, and test images. The training images were transformed and augmented using various methods. The test images were classified to determine the patterns that could be correctly classified. Convolutional neural networks were used for training. Furthermore, we also considered the case where the crystal subcategories were not treated as separate. Learning with all parameters except the random cropping parameter showed the highest accuracy value. Treating the subcategories together or separately did not seem to affect the accuracy value. The accuracy of the best pattern was 0.918. When matched to a real-world case, the percentage of correct answers was 88%. Although the number of images was limited, good results were obtained in the classification of crystal images with optimal parameter tuning. The parameter optimization performed in this study can be used as a reference for future studies, with the goal of image classification by deep learning in clinical practice.

DOI： 10.1038/s41598-022-25385-x

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Schwannomas consist of both high-cellularity regions (Antoni A area) and hypocellular regions (Antoni B area) in histopathological findings. Neurofibromas characteristically consist of CD34 positive spindle cells with thin, wavy, nuclei and wavy collagen bands. Previous reports have described segments of schwannomas with neurofibroma features as hybrid tumors, although hybrid tumors were diagnosed based on partial CD34 positivity in many previous reports. On the other hand, the Antoni B area of some schwannomas was reported to be positive for CD34. Therefore, the definition of a hybrid tumor has not been clear. The objective of this study was to determine whether only CD34 positive findings in schwannomas could be used to define a hybrid tumor. In the analysis of our patient with schwannomatosis caused by a novel LZTR1 germline mutation, part of the tumor had CD34 positive hypocellular regions. These regions contained no thin, wavy, nuclei, indicating an Antoni B area. Laser microdissection was used to investigate the genetic background and differences in molecular mechanisms between CD34 positive and CD34 negative regions. All mutations identified in CD34 positive regions were also found in CD34 negative regions. Our data could not clear the genetic background of Antoni B which was CD34 positive area. We then reviewed the pathologies of 66 sporadic schwannomas. Histopathological examinations of all schwannomas revealed the absence of thin, wavy, nuclei and wavy collagen bands, and no hybrid tumors were found in any of the cases. Ten of 66 patients were randomly selected for CD34 immunostaining and positivity was found in all cases. Our data suggest that it is difficult to distinguish schwannomas by staining for CD34 alone, as Antoni B areas can also be positive for CD34. Therefore, CD34 staining alone should not be used to diagnose hybrid tumors in patients with schwannomas.

DOI： 10.1111/1346-8138.17343

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Elsevier BV  

DOI： 10.1016/j.humpath.2024.02.009

researchmap

記述言語：日本語   出版者・発行元：新潟医学会  

脳の視覚野にヒントを得て考案された深層学習ネットワークによる画像処理技術への医療への応用が期待されている.深層学習は,コンピュータの発展による扱えるデータ量の増大によって発展し,複雑な課題を精度よくこなす技術が開発されている.しかし,深層学習の医療への応用には,データの収集の困難さをはじめとした,いくつかの問題点がある.本稿では,データの前処理で精度向上を試みた著者らの取り組みとともに,深層学習による画像処理技術の発展と今後の展望について紹介する.(著者抄録)

researchmap

記述言語：日本語

CiNii Books

researchmap

記述言語：日本語

CiNii Books

researchmap

記述言語：日本語

CiNii Books

researchmap