基于超声征象的决策树模型在预测胃肠道间质瘤危险度中的应用价值

doi:10.19401/j.cnki.1007-3639.2022.01.005

摘要/Abstract

摘要：

背景与目的：胃肠道间质瘤（gastrointestinal stromal tumor,GIST）的生物危险度可分为极低危、低危、中危和高危,术前就危险度进行准确预判对临床制订诊疗方案至关重要。探讨基于超声征象的决策树模型在预测GIST危险度中的应用价值。方法：收集福建医科大学附属协和医院2016年12月—2020年12月收治的206例GIST患者,建立决策树模型,并判断模型预测的准确度。结果：低危、中危和高危3组间病灶长径、短径/长径（short diameter/long diameter,S/L）、肿瘤部位、回声高低、回声均匀性、边界、形态、是否坏死囊变和血流信号的差异均有统计学意义（P均<0.05）。以病灶长径、肿瘤部位、回声均匀性及形态为节点建立的决策树模型预测的准确度为72.33%,其中低危组和高危组预测的准确度可高达80.90%和93.90%。结论：以超声特征为主要指标构建的决策树模型在预测GIST危险度中具有较高的应用价值。

关键词: 胃肠道间质瘤, 危险分级, 决策树, 超声

Abstract:

Background and purpose: The biological risk of gastrointestinal stromal tumor (GIST) can be divided into very low-risk, low-risk, moderate-risk and high-risk. Accurate preoperative prediction of risk is very important for clinical diagnosis and treatment. This study explored the application value of decision tree model based on ultrasonic signs in predicting the risk grade of GIST. Methods: A total of 206 GIST patients treated in Union Hospital Affiliated to Fujian Medical University from December 2016 to December 2020 were collected. The decision tree model was established, and the prediction accuracy of the model was calculated. Results: There were statistically significant differences in long diameter, short diameter/long diameter (S/L), location, internal echo, echo uniformity, boundary, shape, cystic necrosis and blood flow signals among low-risk, moderate-risk and high-risk groups (all P<0.05). The prediction accuracy of the decision tree model based on the long diameter, location, echo uniformity and shape was 72.33%, and the prediction accuracies of the low-risk group and high-risk group were 80.90% and 93.90%, respectively. Conclusion: The decision tree model based on ultrasonic signs has high application value in predicting the risk grade of GIST.

Key words: Gastrointestinal stromal tumor, Risk grade, Decision tree, Ultrasonography

中图分类号:

郭晶晶, 唐秀斌, 陈蕾, 钱清富, 林礼务, 薛恩生, 陈志奎. 基于超声征象的决策树模型在预测胃肠道间质瘤危险度中的应用价值[J]. 中国癌症杂志, 2022, 32(1): 41-46.

GUO Jingjing, TANG Xiubin, CHEN Lei, QIAN Qingfu, LIN Liwu, XUE Ensheng, CHEN Zhikui. Application value of decision tree model based on ultrasonic signs in predicting the risk grade of gastrointestinal stromal tumors[J]. China Oncology, 2022, 32(1): 41-46.

图/表 5

表 1

表 2

图 1

图 2

表 3

参考文献 19

[1]	NISHIDA T, GOTO O, RAUT C P, et al. Diagnostic and treatment strategy for small gastrointestinal stromal tumors[J]. Cancer, 2016, 122(20): 3110-3118. doi: 10.1002/cncr.30239
[2]	管建明, 茹翱. 有回声型胃超声显像剂对诊断胃间质瘤的价值探讨[J]. 中华超声影像学杂志, 2017, 26(4): 360-361.
	GUAN J M, RU A. The value of echogenic gastric ultrasonography agent in diagnosis of gastric stromal tumors[J]. Chin J Ultrasonogr, 2017, 26(4): 360-361.
[3]	ZHENG T, DU J, YANG L, et al. Evaluation of risk classifications for gastrointestinal stromal tumor using multi-parameter magnetic resonance analysis[J]. Abdom Radiol (NY), 2021, 46(4): 1506-1518.
[4]	LEE M W, KIM G H, KIM K B, et al. Digital image analysis-based scoring system for endoscopic ultrasonography is useful in predicting gastrointestinal stromal tumors[J]. Gastric Cancer, 2019, 22(5): 980-987. doi: 10.1007/s10120-019-00928-w
[5]	LEE H S, CHO C M, KWON Y H, et al. Predicting malignancy risk in gastrointestinal subepithelial tumors with contrast-enhanced harmonic endoscopic ultrasonography using perfusion analysis software[J]. Gut Liver, 2019, 13(2): 161-168. doi: 10.5009/gnl18185
[6]	SMAYRA T, CHARARA Z, SLEILATY G, et al. Classification and regression tree (CART) model of sonographic signs in predicting thyroid nodules malignancy[J]. Eur J Radiol Open, 2019, 6: 343-349. doi: 10.1016/j.ejro.2019.11.003
[7]	PERRIER L, RASCLE P, MORELLE M, et al. The cost-saving effect of centralized histological reviews with soft tissue and visceral sarcomas, GIST, and desmoid tumors: the experiences of the pathologists of the French Sarcoma Group[J]. PLoS One, 2018, 13(4): e0193330. doi: 10.1371/journal.pone.0193330
[8]	VAICEKAUSKAS R, URBONIENĖ J, STANAITIS J, et al. Evaluation of upper endoscopic and endoscopic ultrasound features in the differential diagnosis of gastrointestinal stromal tumors and leiomyomas in the upper gastrointestinal tract[J]. Visc Med, 2020, 36(4): 318-324. doi: 10.1159/000504327
[9]	TAKAHASHI K, NIHEI T, AOKI Y, et al. Gastric gastrointestinal stromal tumor with predominant cystic formation diagnosed by endoscopic ultrasound-fine needle aspiration[J]. Clin J Gastroenterol, 2020, 13(3): 359-364. doi: 10.1007/s12328-019-01058-7
[10]	BAZAGA PÉREZ DE ROZAS S, GALLARDO RAMÍREZ M A, GARCÍA-ALONSO F J, et al. Endoscopic ultrasound-guided radiofrequency ablation for management of gastric gastrointestinal stromal tumor[J]. Endoscopy, 2019, 51(8): E223-E224.
[11]	JOENSUU H. Risk stratification of patients diagnosed with gastrointestinal stromal tumor[J]. Hum Pathol, 2008, 39(10): 1411-1419. doi: 10.1016/j.humpath.2008.06.025
[12]	ZHAO L F, ZHAO Z G, WANG W D, et al. Current characteristics on small intestinal stromal tumor-a case control study[J]. Ann Palliat Med, 2020, 9(1): 98-107. doi: 10.21037/apm
[13]	李清清, 蔡胜. 伴有肠瘘的小肠间质瘤超声表现1例[J]. 中华超声影像学杂志, 2015, 24(2): 158, 163.
	LI Q Q, CAI S. Ultrasonography of small intestinal stromal tumor with intestinal fistula: a case report[J]. Chin J Ultrasonogr, 2015, 24(2): 158, 163.
[14]	LI H, REN G, CAI R, et al. A correlation research of Ki-67 index, CT features, and risk stratification in gastrointestinal stromal tumor[J]. Cancer Med, 2018, 7(9): 4467-4474. doi: 10.1002/cam4.2018.7.issue-9
[15]	ROŞULESCU A, PECHIANU N, HORTOPAN M, et al. Gastrointestinal stromal tumors of the colon and rectum[J]. Pol J Pathol, 2020, 71(3): 200-206.
[16]	王子千, 丁晓玲, 曹颖, 等. 小肠高危间质瘤坏死囊性变并窦道形成超声表现3例[J]. 中华超声影像学杂志, 2013, 22(4): 324, 328.
	WANG Z Q, DING X L, CAO Y, et al. Ultrasonographic findings of high-risk small intestinal stromal tumor with necrotic cystic and sinus tract: 3 cases report[J]. Chin J Ultrasonogr, 2013, 22(4): 324, 328.
[17]	NISHIDA T, SAKAI Y, TAKAGI M, et al. Adherence to the guidelines and the pathological diagnosis of high-risk gastrointestinal stromal tumors in the real world[J]. Gastric Cancer, 2020, 23(1): 118-125. doi: 10.1007/s10120-019-00966-4
[18]	SASMAL P K, SHARMA R, PATRA S, et al. Malignant extra-gastrointestinal stromal tumor of the mesentery[J]. Surg J (N Y), 2019, 5(3): e65-e68.
[19]	PESENTI C, BORIES E, CAILLOL F, et al. Characterization of subepithelial lesions of the stomach and esophagus by contrast-enhanced EUS: a retrospective study[J]. Endosc Ultrasound, 2019, 8(1): 43-49. doi: 10.4103/eus.eus_89_17

Indicator	Low-risk group vs moderate-risk group		Moderate-risk group vs high-risk group		Low-risk group vs high-risk group
Indicator	Statistic of test	P value	Statistic of test	P value	Statistic of test	P value
Long diameter^b	-13.062	0.240	-61.837	0.000	-74.900	0.000
S/L^b	-0.837	0.944	27.516	0.022	26.679	0.003
Tumor source location^a	8.824	0.004	21.315	0.000	8.991	0.003
Internal echo^c	NA	0.006	NA	0.350	NA	0.050
Echo uniformity^a	20.959	0.000	1.562	0.301	52.159	0.000
Boundary	NA	0.466^c	3.366^a	0.089^a	12.577^a	0.000^a
Shape^a	7.656	0.009	4.960	0.030	37.611	0.000
Necrotic cystic degeneration^a	14.100	0.000	5.362	0.025	51.831	0.000
Blood flow signals^b	4.365	0.696	-60.184	0.000	-55.819	0.000

Prediction	Pathology
Prediction	Low-risk	Moderate-risk	High-risk	Total
Low-risk	72	12	5	89
Moderate risk	0	0	0	0
High-risk	17	23	77	117
Total	89	35	82	206