http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : ( Harsha Pokkalla ) (검색결과 1 건)
- 무료
- 기관 내 무료
- 유료
-
( Harsha Pokkalla ),( Kishalve Pethia ),( Benjamin Glass ),( Jennifer Kaplan Kerner ),( Ling Han ),( Catherine Jia ),( Ryan Huss ),( Mar-ianne Camargo ),( Kathryn Kersey ),( Chuhan Chung ),( G. Mani S 대한간학회 2020 춘·추계 학술대회 (KASL) Vol.2020 No.1
Aims: Fibrosis is the primary determinant of disease progression in patients with nonalcoholic steatohepatitis (NASH), but the prognostic impact of other histological features is unclear. We used a machine learning(ML) approach to identify novel morphologic features and associations with disease progression in NASH patients with F3/4 fibrosis. Methods: Biopsies from 644 patients screened in phase3 trial of selonsertib (STELLAR-4) were scored by a central pathologist( CP) according to the NASH CRN and Ishak staging systems. The PathAI research platform(PathAI, Boston, MA) was trained a convolutional neural network(CNN) with >68,000 annotations (e.g. steatosis, ballooning, lobular/portal inflammation) collected from 75 board-certified pathologists on images of H&E and trichrome(TC) stained slides. For staging fibrosis, CNN models were trained using slide-level pathologist scores to recognize unique patterns associated with each stage within fibrotic regions of TC images. 202 features were extracted from biopsy images from patients (F3-F4) enrolled in the STELLAR trials. Cox regression was used to identify associations between these features with progression to cirrhosis in F3 patients, and liver-related events (e.g. decompensation, transplantation, death) in F4 patients. Results: 1526 NASH patients with F3-F4 fibrosis (median age 59 yrs, 73% diabetic, 52% F4) were included. During a median follow-up of 16.5 mos, 14.5% (105/726) of F3 patients progressed to cirrhosis, and over 15.9 mos, 2.8% (22/800) of F4 patients had liver-related events. Progression to cirrhosis was associated with greater area of Ishak 6 fibrosis and portal inflammation (Figure). Similar associations were observed in F4 patients, with hepatocellular ballooning and clinical events. In F3, a greater proportion of area of Ishak 1 fibrosis and steatosis were associated with a reduced risk of progression. In F4, area of steatosis was similarly protective, while proportion of Ishak Stage 1 Fibrosis over Ishak scored area trended towards protective. Conclusions: Liver histological evaluation using ML approach identified novel features associated with progression in NASH patients with advanced fibrosis. These data support the utility of ML approaches to evaluation of liver histology as endpoints in NASH clinical trials.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
