自動傷口判讀及術後傷口追蹤評估系統

Infection Detection Method and Surgical Wound Assessment System

徐瑞澤(Jui-Tse Hsu)

指導教授 ： 賴飛羆

國立臺灣大學/電機資訊學院/生醫電子與資訊學研究所/博士(2019年)

https://doi.org/10.6342/NTU201903142

摘要

截至目前為止，針對開完刀之患者傷口進行照護之工作，不管是對醫護人員還是病人本身，都還是一個既繁瑣又相當具有挑戰性的工作，為了減輕醫護人員以及病人的負擔，本研究發展了一系列的方法及一套完整的系統，用以自動偵測術後傷口位置，並評估傷口是否出現感染跡象。第一部分之研究主要分為以下兩個部分：(1) 全自動傷口影像切割演算方法，以及(2) 機器學習演算法以判讀傷口之感染情形。針對(1)，此部分利用手機或相機等手持式裝置對傷口進行拍照後，接著使用此研究所發展出之數位影像分析技術，自動從傷口影像中偵測並切割出傷口區域。此部分開發完成的技術包括： A. 強健邊緣檢測，B. 膚色區域劃分，C. 傷口區域劃分，及D. 傷口區域重建。針對(2)則發展了一套基於機器學習演算方法所建立之全自動傷口分析演算法，此部分會利用前述第一步驟所切割出的傷口區域，從中抓取縫合位置及特徵點，並透過本研究所提出的最佳化分群方法將這些特徵點形成多個集合，並合併這些集合形成多個感興趣區域 ROI (Region of Interest)，每一個ROI分別代表不同的傷口縫合位置，接著從這些ROI區域建立特徵向量並尋找傷口感染之跡象。此部分開發完成的技術包括： A. 傷口縫合處定位，B. 感興趣區域偵測，C. 特徵選取，及D. 使用機器學習訓練傷口辨識器。
本研究之第二部分修正了第一部份方法採用閥值進行影像切割的作法，採用機器學習的方法判斷背景影像資訊，並提出一套完整的術後傷口追蹤評估系統架構，及與之搭配之傷口照護APP，以利患者及醫護人員拍攝傷口後上傳至系統進行全自動傷口追蹤感染分析。此系統使用之演算方法可以分為以下四個部分：第一部份，傷口影像區域首先會依照相似度進行超像素切割；第二部分，辨認屬於皮膚之超像素，並正確決定出影像中之皮膚區域；第三部分，根據傷口與皮膚之材質差異，正確的傷口區域會被辨認出來；最後，針對每一個傷口區域，系統會針對傷口正常與否，以及是否發生任何之感染症狀以進行分析評估。

關鍵字

群聚 ； 邊緣偵測 ； 影像切割 ； 機器學習 ； 人工智慧 ； 醫學影像處理 ； 傷口分類 ； 傷口評估 ； 感染評估

並列摘要

Numerous patients suffer from chronic wounds and wound infections nowadays. Until now, the care for wounds after surgery still remain a tedious and challenging work for the medical personnel and patients. As a result, with the help of the hand-held mobile devices, there is high demand for the development of a series of algorithms and related methods for wound infection early detection and wound self-monitoring.
This research proposed an automated way to perform (1) wound image segmentation and (2) wound infection assessment after surgical operations. The first part describes an edge-based self-adaptive threshold detection image segmentation method to exclude nonwounded areas from the original images. The second part describes a wound infection assessment method based on machine learning approach. In this method, the extraction of feature points from the suture area and an optimal clustering method based on unimodal Rosin threshold algorithm that divides feature points into clusters are introduced. These clusters are then merged into several regions of interest (ROIs), each of which is regarded as a suture site. Notably, a support vector machine (SVM) can automatically interpret infections on these detected suture site.
This research also propose a surgical wound assessment system. The proposed system is designed to enable patients capture surgical wound images themselves by using a mobile device and upload these images for analysis. Combining image-processing and machine-learning techniques, the proposed method is composed of four phases. First, images are segmented into superpixels where each superpixel contains pixels with similar color distribution. Second, these superpixels corresponding to the skin are identified and the area of connected skin superpixels is derived. Third, surgical wounds will be extracted from this area based on the observation of the texture difference between skin and wounds. Lastly, state and symptoms of surgical wound will be assessed. In this research, I had developed a wound analysis APP to allow patients and medical professionals to capture the wound area and upload those images onto the system for further analysis.

並列關鍵字

Clustering ； Edge Detection ； Image Segmentation ； Machine Learning ； Medical Image Processing ； Surgical Site Classification ； Wound Assessment

參考文獻

[1] Sen, C.K., Gordillo, G.M., Roy, S., Kirsner, R., Lambert, L., Hunt, T.K., Gottrup, F., Gurtner, G.C., and Longaker, M.T., Human skin wounds: A major and snowballing threat to public health and the economy. Wound. Repair. Rege.. 17(6):763–771, 2009.

[2] V. Dini, P. Salvo, A. Janowska, F. Di Francesco, A. Barbini, and M. Romanelli, "Correlation Between Wound Temperature Obtained With an Infrared Camera and Clinical Wound Bed Score in Venous Leg Ulcers," Wounds: a compendium of clinical research and practice, vol.27, no.10, pp.274-278, 2015.

[3] D. Lubeley, K. Jostschulte, R. Kays, K. Biskup, and B. Clasbrummel, "3D wound measurement system for telemedical applications," In 39th Annual Congress of the German Society for Biomedical Engineering, ISSN, pp.0939-4990, 2005.

[4] A. F. M. Hani, N. M. Eltegani, S. H. Hussein, A. Jamil, and P. Gill, "Assessment of ulcer wounds size using 3D skin surface imaging," In Visual Informatics: Bridging Research and Practice, pp.243-253, 2009.

[5] H. Wannous, Y. Lucas, and S. Treuillet, "Enhanced assessment of the wound-healing process by accurate multiview tissue classification," Medical Imaging, IEEE Transactions on 30.2, pp.315-326, 2011.