臺灣博碩士論文加值系統

在美國每年約有100萬人因頭部外傷進行頭部電腦斷層檢查，這其中輕度頭部外傷佔了約80%。對到達急診時昏迷指數(Glasgow Coma Scale, GCS) 15分的病人而言，約有3-15%電腦斷層檢查是有異常的，其中只有1-5%是需要神經外科手術治療的。
本研究的資料來源是從「中華民國頭脊髓外傷研究小組」所收集台灣全國頭部外傷資料庫，並從中挑選出自民國94年7月1號到民國95年6月30號，這其間發生輕度頭部外傷之傷病人來做研究，一共有2,294人。利用頭部外傷發生的機轉如車禍，跌落，墜落物襲擊及患者所使用的交通工具與相撞之物體等及臨床資料如年齡、性別、有顱骨骨折、局部神經功能障礙、有失去意識、創傷後癲癎、失憶、等來建構一類神經網路預測模型用來預測輕部頭部外傷後電腦斷層是否出現異常。最後，我們利用ROC 曲線(receiver operating characteristic curve)及ROC分析法對此一神經網預測模型之鑒別能力作分析，也將此分析結果與傳統统計學邏輯迴歸模型作比較。
在我們的研究中，大約有44.2%的輕度頭部外傷後電腦斷層出現異常，而我們所建構的類神經網路模式為一個含有17個輸入變項的多層感知機網路。測試結果發現類神經網路模式有最大的ROC曲線下方面積(AUC：0.887)。類神經網路模式與邏輯迴歸模式之AUC無統計學上之顯著差異。然而，在兩個關鍵的ROC曲線臨界點上類神經網路預測模式都有最佳之共存敏感度及特異度。若把2組敏感度固定為90.3%，則類神經網路特異度為比邏輯廻歸的表現好(87% v.s. 50.3%)，且達到顯著差異。
未來如果能結合臨床上用以預測輕度頭部外傷後電腦斷層出現更多的臨床變項（如嘔吐、頭痛等）納入建構類神經網路模式，相信類神網路將可協臨床醫師在做決定讓病人進行電腦斷層與否多一層把握。

About one million people in the US receive brain computed tomographic examination every year due to head injuries.with 80% of them suffering a mild head injury. Among patients arriving in the emergency room with a Glasgow Coma Scale (GCS) of 15, 3%-15% of them were found to be abnormal in their brain CT scan, and up to 5% of them required neurosurgical intervention.
We used the Taiwan National Head Injury Database collected by the Head and Spinal Injury Research Team. we selected patients (n=2,294) with a mild head injury in from July 1,2005 to June 30,2006. Based on information from the database and clinical research material, we then established a neural network to predict whether any abnormality appeared in the brain CT in patients with a mild head injury. The ANN’S ability to discriminate outcomes was assessed using receiver operating characteristic (ROC) curve analysis and the results were compared with a multiple logistic regression model.
The results of the study that brain CT abnormalities following a mild head injury occurred in 44.2% of the patients, and that 17 predictive variables were identified by the ANN model. When tested on the same validation set, the ANN model had the greatest area under the ROC curve (AUC=0.887). we also found that a pair-wise comparison between the ANN and LR (AUC=0.861) models showed a similar predictive performance (p=0.078). The ANN model had the best simultaneous sensitivity and specificity at two pre-defined cut-off values. If the cut-off point sensitivity was fixed at an optimal level of 90.3%,we found the study showed that the ANN model had a better specificity and performance than a classical statistical logistical regression model (87% vs. 50.3%)
Based on this study results, we conclude that we can merge clinical tests to determinate the exceptional outliers from the CT results after studying mild head injury, symptoms, such as vomiting and headache, to help clinicians better decide on whether to order brain CT scan.

第一章 緒論 1
第一節 研究背景 1
第二節 研究目的 5
第三節 名詞定義 6
第二章 文獻探討 10
第一節輕度頭部外傷之定義 10
第二節 預測頭部外傷後腦出血或電腦斷層異常的危險因子 13
第三節 以類神經網路模式來預測輕度頭部外傷後電腦斷層異常 18
第三章 研究方法與步驟 21
第一節 研究對象 21
第二節 研究地區與時間 24
第三節 研究架構 26
第四節 變數的選取 28
第五節 類神經網路模式 30
第六節 邏輯迴歸模式 41
第四章 結果與分析 43
第五章 討論 63
第六章 結論 72
第七章 限制 73
參考文獻 75

1.McCaig LF, Ly N. National Hospital Ambulatory Medical Care Survey: 2000 Emergency Department Summary. Adv Data. 2001; 326:1-27.
2.Borczuk P. Predictors of intracranial injury in patients with mild head injury. Ann Emerg Med. 1995;25:731-736.
3.Thurman DJ, Alverson C,Dunn KA. Traumatic Brain Injury in the United States: A public health perspective.J Head Trauma Rehabil 14(6):602,1999.
4.Finkelstein E et al. The Incidence and Economic Burden of Injuries in the United States. New York: Oxford University Press, 2006.
5.Jagoda AS, Cantrill SV, Wears RL .Clinical Policy: Neuroimaging and Decisionmaking in Adult Mild Traumatic Brain Injury in the Acute Setting.Ann Emerg Med .August 2002;40:231-249.
6.Nagy KK, Joseph KT, Krosner SM, et al. The utility of head computed tomography after minimal head injury. J Trauma. 1999;46:268-270.
7.Stein SC, Ross SE. Mild head injury: a plea for routine early CT scanning. J Trauma. 1992;33:11-13.
8.Hofman PAM, Nelemans P, Kemerink GJ, et al. Value of radiological diagnosis of skull fracture in the management of mild head injury: meta-analysis. J Neurol Neurosurg Psychiatry. 2000;68:416-422.
9.Haydel MJ, Preston CA, Mills TJ, et al. Indications for computed tomography in patients with minor head injury. N Engl J Med. 2000;343:100-105.
10.Miller EC, Derlet RW, Kinser D. Minor head trauma: is computed tomography always necessary? Ann Emerg Med. 1996;27:290-294.
11.Advanced Trauma Life Support Program for Doctor 6th Edition 1997.chapter 6:205-213.
12.Teasdate G, Jennett B: Assessment of coma and impaired conciousness: Lance,1974.
13.Servadei F, Teasdale G, Merry G, Neurotraumatology Committee of the World Federation of Neurosurgical Societies: Defining acute mild head injury in adults: A proposal based on prognostic factors, diagnosis, and management. J Neurotrauma 18(7):657,2001.
14.Thomas DK, Christopher AL :Head injury. In Tintinalli JE,et al (eds):Emergency Medicine: A Comprehensive Study Guide.6th ed. New York, McGraw-Hill, 1157-1569.
15.Iverson GL, Lovell MR, Smith S, Franzen MD: Prevalence of abnormal CT scans following mild head injury. Brain Inj 14(12):1057,2000.
16.Livingston DH, Loder PA, Koziol J, et al. The use of CT scanning to triage patients requiring admission following minimal head injury. J Trauma. 1991;31:483-489.
17.Shackford SA, Wald SL, Ross SE, et al. The clinical utility of computed tomographic scanning and neurologic examination in the management of patients with minor head injuries. J Trauma. 1992;33:385-394.
18.Saunders CE, Cota R, Barton CA. Reliability of home observation for victims of mild closed head injury. Ann Emerg Med. 1986;15:160-163.
19.Levitt MA, Sutton M, Goldman J, et al. Cognitive dysfunction in patients suffering minor head trauma. Am J Emerg Med.1994;12:172-174.
20.Krause JF et al. Epidemiology of brain injury.In Narayan RK, Wilberger WK, Povlishock JT. (eds): Neurotrauma, McGraw-Hill 1995 ,p13.
21.Williams DH, Levin HS, Eisenberg HM. Mild head injury classification. Neurosurgery.1990;27:422-428.
22.Lee ST, Liu TN, Wong CW, et al. Relative risk of deterioration after mild closed head injury. Acta Neurochir (Wien). 1995;135:136-140.
23.Dietrich AM, Bowman MJ, Ginn-Pease ME, Kosnik E, King DR.Pediatric head injuries: can clinical factors reliably predict an abnormality on computed tomography? Ann Emerg Med. 1993;22:1535–1540.
24.Ramundo ML, McNight T, Kempf J, Satkowiak L. Clinical predictors of computed tomographic abnormalities following pediatric traumatic brain injury. Pediatr Emerg Care. 1995;11:1–4.
25.Madden C, Witzke DB, Sanders AB, et al. High-yield selection criteria for cranial computed tomography after acute trauma. Acad Emerg Med. 1995;2:248-253.
26.Miller EC, Derlet RW, Kinser D. Minor head trauma: is computed tomography always necessary? Ann Emerg Med. 1996;27:290-294.
27.National center for health statistics.National hospital ambulatory medical care survey 1992.Hyattsville: National center for health statistics 1994.
28.Steven Mandel et al. Minor Head Trauma Assessment, management, and Rehabilitation 1993. chapter1:1-4, chapter7,8:107-141.
29.劉建雄，邱文達，林烈生，洪慶章：外傷指數及外傷嚴重度指數應用於國內外傷病患之初報：兩所醫院外科急診1000病例分析。中華醫學雜誌。1991，48(4)：297-304。
30.Kraus JF：The epidemiology relevance of the medical literature published from 1960-1978.J Neurosurg.1980,53:803-810.
31.Annegers JF, Grabow JD et al: The incidence , cause, and secular trends of head trauma in Olmestedt country, Minnesota,1935-1974.Neurology 1980, 30: 912-919.
32.Advanced Trauma Life Support Program for Doctor. 6th Edition 1997.Chapter 6. 205-213.
33.Min-Huei Hsu et al. Predicting cranial computed tomography results of head injury patients using an artificial neural network AMIA Annu Symp Proc. 2003;:868.
34.Victor G. MD, MPH; Thomas, Karen E. MPH; Sattin, Richard W.MD et al .The CDC Traumatic Brain Injury Surveillance System: Characteristics of Persons Aged 65 Years and Older Hospitalized With a TBIJ Head Trauma Rehabil Vol. 20, No. 3, pp. 215–228c 2005 Coronado.
35.Kosakevitch-Ricbourg L et al. Mild brain injuries: definition, classifications and prognosis. Rev Stomatol Chir Maxillofac. 2006 Sep;107(4):201-5. French
36.Pickett W, Ardern C, Brison RJ. et al CMAJ. A population-based study of potential brain injuries requiring emergency care. 2001 Aug 7;165(3):288-92.
37.Martin RC 2nd, Spain DA, Richardson JD. et al. Do facial fractures protect the brain or are they a marker for severe head injury? Am Surg. 2002 May;68(5):477-81.
38.Dunham CM, Coates S, Cooper C. Compelling evidence for discretionary brain computed tomographic imaging in those patients with mild cognitive impairment after blunt trauma. J Trauma.1996;41:679-686.
39.Vilke GM, Chan TC, Guss DA. Use of a complete neurological examination to screen for significant intracranial abnormalities in minor head injury. Am J Emerg Med. 2000;18:159-163.
40.Stiell IG, Wells GA, Vandeheen K,et al. The Canadian CT Head Rule for patients with mild head injury. Lancet. 2001;357:213-219.
41.Marion Smits, Diederik W, Gijs G, et al.External Validation of the Canadian CT Head Rule and the New Orleans Criteria for CT scanning in Patients With Minor Head Injury. JAMA.2005;294:1519-1525.
42.William RM, Jerome RH, Mel Herbert, et al. Developing a decision instrument to guide computed tomographic imaging of blunt head injury patients .J Trauma.2005; 59:954-959.
43.Forsstrom JJ, Dalton KJ. Artificial neural networks for decision support in clinical medicine. Ann Med.1995;27:509-517.
44.Yu-Chuan Li, Li Liu, Wen-Ta Chiu, et al.Neural network modeling for surgical decisions on traumatic brain injury patients.International J of Med Inform.57(2000)1-9.
45.Min-Huei Hsu, Yu-Chuan Li,wen-Ta Chiu,et al. Predicting cranial computed tomography results of head injury patients using an artificial neural network AMIA Annu Symp Proc. 2003:868.
46.Madhumita Sinha, Christopher SK, Maria LR. Artificial Neural Network predicts CT scan abnormalities in pediatric patients with closed head injury. J Trauma.2001;50:308-312.
47.Joel D, Phil SS, Fiona L, et al. A Meta-Analysis of Clinical Correlates that Predict Significant Intracranial Injury in Adults with Minor Head Trauma. J of Neurotrauma. 2004;21(7):877-885.
48.Mark HZ, Gregory C. Receiver-Operating Characteristic (ROC) Plots: A Fundamental Evaluation Tool in Clinical Medicine .CLINICAL CHEMISTRY.1993;39(4):561-577.
49.Swets JA. Measuring the accuracy of diagnostic systems. Science 1988;18:229-235.
50.Sargent DJ: Comparison of artificial neural networks with other statistical approaches. Cancer (supplement) 2001,91(8):636-642.
51.Behzad E, Kazem M, Hassan EA et al. Comparison of artificial neural networks and logistic regression models for prediction of mortality in head trauma based on initial clinical data. BMC med Inf and Decision making. 2005:5:3:1-8.
52.Tu JV. Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. J Clinical Epidemiologic .1996;49:1225-1231.
53.Selker HP, Griffith JL, Patil S et al. A comparison of performance of mathematical predictive methods for medical diagnosis:identifying acute ischema among emergency department patients. J Investig Med.1995;43:468-476.