Faster RCNN和LGDF結(jié)合的肝包蟲病CT圖像病灶分割
2021年電子技術(shù)應(yīng)用第7期
劉志華1,,王正業(yè)1,,李豐軍2,嚴(yán)傳波2
1.新疆醫(yī)科大學(xué) 公共衛(wèi)生學(xué)院,,新疆 烏魯木齊830011,;2.新疆醫(yī)科大學(xué) 醫(yī)學(xué)工程技術(shù)學(xué)院,,新疆 烏魯木齊830011
摘要: 針對(duì)人工閱片工作量大、閱片質(zhì)量不佳且容易出現(xiàn)漏檢,、錯(cuò)判等問題,,將Faster RCNN目標(biāo)檢測模型應(yīng)用于肝包蟲病CT圖像的檢測,并對(duì)目標(biāo)檢測模型進(jìn)行改進(jìn):基于圖片分辨率低,、病灶大小不同的特點(diǎn),,使用網(wǎng)絡(luò)深度更深的殘差網(wǎng)絡(luò)(ResNet101)代替原來的VGG16網(wǎng)絡(luò),用以提取更豐富的圖像特征,;根據(jù)目標(biāo)檢測模型得出的病灶坐標(biāo)信息引入LGDF模型進(jìn)一步對(duì)病灶進(jìn)行分割,,從而輔助醫(yī)生更高效的診斷疾病。實(shí)驗(yàn)結(jié)果表明,,基于ResNet101特征提取網(wǎng)絡(luò)的目標(biāo)檢測模型能夠有效提取目標(biāo)的特征,,檢測準(zhǔn)確率相比原始檢測模型提高2.1%,具有較好的檢測精度,。同時(shí),,將病灶坐標(biāo)信息引入LGDF模型,相比于原始的LGDF模型更好地完成了對(duì)肝包蟲病病灶的分割,,Dice系數(shù)提高了5%,,尤其對(duì)多囊型肝包蟲病CT圖像的分割效果較好,。
中圖分類號(hào): TN911.73;TP751.1
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.200923
中文引用格式: 劉志華,,王正業(yè),,李豐軍,等. Faster RCNN和LGDF結(jié)合的肝包蟲病CT圖像病灶分割[J].電子技術(shù)應(yīng)用,,2021,,47(7):33-37,43.
英文引用格式: Liu Zhihua,,Wang Zhengye,,Li Fengjun,et al. CT image segmentation of liver hydatid disease based on Faster RCNN and LGDF[J]. Application of Electronic Technique,,2021,,47(7):33-37,43.
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.200923
中文引用格式: 劉志華,,王正業(yè),,李豐軍,等. Faster RCNN和LGDF結(jié)合的肝包蟲病CT圖像病灶分割[J].電子技術(shù)應(yīng)用,,2021,,47(7):33-37,43.
英文引用格式: Liu Zhihua,,Wang Zhengye,,Li Fengjun,et al. CT image segmentation of liver hydatid disease based on Faster RCNN and LGDF[J]. Application of Electronic Technique,,2021,,47(7):33-37,43.
CT image segmentation of liver hydatid disease based on Faster RCNN and LGDF
Liu Zhihua1,,Wang Zhengye1,,Li Fengjun2,Yan Chuanbo2
1.College of Public Health,,Xinjiang Medical University,,Urumqi 830011,China,; 2.College of Medical Engineering Technology,,Xinjiang Medical University,Urumqi 830011,,China
Abstract: In view of the large workload of manual image reading, poor image reading quality, and prone to missed inspections and wrong judgments,,in this paper, the faster RCNN target detection model is applied to the detection of hepatic echinococcosis CT images. And the target detection model is improved: based on the characteristics of low image resolution and different lesion sizes, the residual network with deeper network depth(ResNet101) is used to replace the original VGG16 to extract richer image features; according to the coordinate information of the lesion obtained by the object detection model, the LGDF model is introduced to further segment the lesion to assist doctors in diagnosing the disease more efficiently. The experimental results show that the object detection model based on the ResNet101 feature extraction network can effectively extract the features of the target, and the detection accuracy is 2.1% higher than the original detection model, and it has better detection accuracy. At the same time, the coordinate information of the lesion is introduced into the LGDF model. Compared with the original LGDF model, the segmentation of hepatic hydatid lesions is better completed, the Dice coefficient is increased by 5%, and the segmentation effect is better especially for the multi cystic liver hydatidosis CT image.
Key words : faster RCNN;LGDF,;deep learning,;object detection;lesion segmentation
0 引言
肝包蟲病(Hepatic Echinococcosis,,HE)又稱棘球幼病,,是一種人畜共患寄生蟲病,主要流行于畜牧業(yè)發(fā)達(dá)地區(qū)[1-3],。肝包蟲病患者在患病初期無特異性的癥狀及體征,,隨著包囊的生長,患者出現(xiàn)臨床癥狀,,引起自身機(jī)體的感染并發(fā)生一些并發(fā)癥,,其中部分并發(fā)癥可能危及患者生命,需要醫(yī)生的及時(shí)診斷和緊急干預(yù)[4-5],。醫(yī)學(xué)影像學(xué)檢查是診斷疾病的一種方式,,能夠?yàn)榛颊叩牟∏樘峁┯杏玫男畔?,?duì)于肝包蟲病的影像學(xué)診斷是由醫(yī)生查看拍攝的CT圖片診斷患者是否發(fā)生疾病。隨著影像設(shè)備的更新和發(fā)展,,醫(yī)院每天產(chǎn)出大量的醫(yī)學(xué)圖片,,醫(yī)生閱片時(shí)容易發(fā)生視覺疲勞現(xiàn)象,往往出現(xiàn)診斷效率低下,、漏檢,、誤判等問題。因此,,本文基于目標(biāo)檢測方法實(shí)現(xiàn)肝包蟲病病灶的檢測,,從而輔助醫(yī)生智能診斷疾病。
本文詳細(xì)內(nèi)容請(qǐng)下載:http://forexkbc.com/resource/share/2000003650,。
作者信息:
劉志華1,王正業(yè)1,,李豐軍2,,嚴(yán)傳波2
(1.新疆醫(yī)科大學(xué) 公共衛(wèi)生學(xué)院,新疆 烏魯木齊830011,;2.新疆醫(yī)科大學(xué) 醫(yī)學(xué)工程技術(shù)學(xué)院,,新疆 烏魯木齊830011)
此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載,。