之前的研究顯示，早期血管內溶栓術（EVT）可以改善急性缺血性中
風（AIS）患者在手術後的結果，特別是那些有大血管阻塞（LVO）的患
者。從2014年開始，EVT已成為對於有LVO的AIS患者的確定性治療方法。目
前，EVT資源，包括設備和人員，僅在具備EVT能力的中風醫院中提供，而
非EVT能力的中風醫院則只能為沒有LVO的中風患者提供確定性治療，即血
栓溶解劑（rt-PA）。通常在醫院內無法在磁共振成像（MRI）之前確認是否
為AIS-LVO患者，因此在現場對疑似中風患者進行醫院選擇通常依賴於院前中
風評估量表。
本研究討論了兩種方法進行醫院選擇。這兩種方法都利用了院前中風評估
量表。第一種方法是標準技術，根據歷史數據使用統計學來確定院前中風評
估量表的最佳切割值。然後，如果現場的患者得分高於最佳切割值，患者將
被送往最近的具備EVT能力的醫院。第二種方法是使用優化模型，將患者的
院前中風評估量表分數轉換為患者患有AIS-LVO的機率度量。可以計算出從發
病到開始確定性治療的預期時間最短的接收醫院。我們使用台北市2019年1月
至2020年12月的中風患者數據來驗證這兩種方法。我們將探討中風病患在不同
的CPSS數值下，分別適合哪些的送醫選擇方案，而我們預期模型的表現應當優
於標準統計方法以及其他的送醫策略。

Previous studies had shown that early endovascular
thrombectomy (EVT) can benefit the outcome of the acute ischemic stroke (AIS) patients with large vessel occlusion (LVO) after the operation. Since year 2014, EVT had become the definitive treatment for the AIS patients with LVO. Currently, the EVT resources including the equipment and personnel are available only in the EVT-capable stroke hospitals but not in the non-EVT-capable stroke hospitals, where the stroke patients without LVO can still receive their definitive treatment, i.e., the recombinant tissue plasminogen activator (rt-PA). Usually, the AIS-LVO patients cannot be confirmed before the magnetic resonance imaging (MRI) in a hospital. Thus, the selection of hospital for the suspected stroke patients on scene often relies on the pre-hospital stroke scales.

This study discusses two methods to do the hospital selection. Both methods make uses of the pre-hospital stroke scales. The first method is the standard technique that uses statistics to determine the optimal cutoffs of the pre-hospital stroke scale based on the historical data. Then, if the patients on scene is with a higher score than that optimal cutoff, the patient will be sent to the nearest EVT-capable hospital. The second method is the technique that uses an optimization model where the patient's score of the pre-hospital stroke scale is converted to a probability measure that the patient is with AIS-LVO. A receiving hospital which minimizes the expected time from the onset to the start of the definitive treatment can be computed. The data of stroke patients between January 2019 and December 2020 in Taipei City is used to validate the two methods. We will explore which strategy is appropriate to the stroke patients at different CPSS (Cincinnati Prehospital Stroke Scale) scores. We anticipate that our model's performance will better than standard statistical methods and the other hospital selection strategies.

Abstract I
Abstract (Chinese) II
Contents III
List of Figures V
List of Tables VI

1 Introduction 1

2 Literature Review 3
2.1 Approach for Assessing the Severity of Stroke 3
2.2 Time sensitivity for the AIS patient 4
2.3 Comparison of Hospital Selection Strategy 4

3 Methodology 6
3.1 Data 6
3.2 Statistical Method 7
3.3 Optimal Hospital Selection Method 11
3.3.1 Notation of Mathematical Model 11
3.3.2 Hospital Selection Model 13
3.4 Strategy 16

4 Result 20

5 Conclusion 26

Bibliography 29

Vicky Chalos, Rob A van de Graaf, Bob Roozenbeek, Adriaan CGM van Es,
Heleen M den Hertog, Julie Staals, Lukas van Dijk, Sjoerd FM Jenniskens, Robert J van Oostenbrugge, Wim H van Zwam, et al. Multicenter randomized clinical trial of endovascular treatment for acute ischemic stroke. the effect of periprocedural medication: acetylsalicylic acid, unfractionated heparin, both, or neither (mr clean-med). rationale and study design. Trials, 21:1–17, 2020.

Jonathan Emberson, Kennedy R Lees, Patrick Lyden, Lisa Blackwell, Gregory Albers, Erich Bluhmki, Thomas Brott, Geoff Cohen, Stephen Davis, Geoffrey Donnan, et al. Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. The Lancet, 384(9958):1929–1935, 2014.

Mayank Goyal, Bijoy K Menon, Wim H Van Zwam, Diederik WJ Dippel,
Peter J Mitchell, Andrew M Demchuk, Antoni D ́avalos, Charles BLM Majoie, ad van Der Lugt, Maria A De Miquel, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. The Lancet, 387(10029):1723–1731, 2016.

Werner Hacke, Markku Kaste, Erich Bluhmki, Miroslav Brozman, Antoni
D´avalos, Donata Guidetti, Vincent Larrue, Kennedy R Lees, Zakaria Medeghri, Thomas Machnig, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. New England journal of medicine, 359(13):1317– 1329, 2008.

Jessalyn K Holodinsky, Tyler S Williamson, Andrew M Demchuk, Henry
Zhao, Luke Zhu, Michael J Francis, Mayank Goyal, Michael D Hill, and
Noreen Kamal. Modeling stroke patient transport for all patients with suspected large-vessel occlusion. JAMA neurology, 75(12):1477–1486, 2018.

Jessalyn K Holodinsky, Tyler S Williamson, Noreen Kamal, Dhruv Mayank,
Michael D Hill, and Mayank Goyal. Drip and ship versus direct to comprehensive stroke center: conditional probability modeling. Stroke, 48(1):233–238, 2017.

Ting-Yu Liu, Chun-Han Wang, Wen-Chu Chiang, Sung-Chun Tang, Li-Kai
Tsai, Chung-Wei Lee, Jiann-Shing Jeng, Matthew Huei-Ming Ma, Ming-Ju
Hsieh, and Yu-Ching Lee. Redistributing medical resources for a bypass strategy for large vessel occlusion: a community-based study. Journal of NeuroIn-terventional Surgery, 12(1):98–103, 2020.

Rachael L MacIsaac, Charith Cooray, Heidi Sucharew, Dawn Kleindorfer,
Pooja Khatri, Joseph P Broderick, Heinrich J Audebert, Niaz Ahmed, Nils Wahlgren, and Matthias Endres. Clinical selection strategies to identify ischemic stroke patients with large anterior vessel occlusion. 2017.

Aditya Maddali, Farook Abdul Razack, Srihari Cattamanchi, and Trichur V. Ramakrishnan. Validation of the cincinnati prehospital stroke scale. Journal of Emergencies, Trauma, and Shock, 11, 2018.

Konark Malhotra, Jeffrey Gornbein, and Jeffrey L Saver. Ischemic strokes due to large-vessel occlusions contribute disproportionately to stroke-related dependence and death: a review. Frontiers in neurology, 8:651, 2017.

Matthew SW Milne, Jessalyn K Holodinsky, Michael D Hill, Anders Nygren, Chao Qiu, Mayank Goyal, and Noreen Kamal. Drip ‘n ship versus mothership for endovascular treatment: modeling the best transportation options for optimal outcomes. Stroke, 48(3):791–794, 2017.


Maxim JHL Mulder, Ivo GH Jansen, Robert-Jan B Goldhoorn, Esmee Ven-
ema, Vicky Chalos, Kars CJ Compagne, Bob Roozenbeek, Hester F Lingsma,
Wouter J Schonewille, Ido R Van Den Wijngaard, et al. Time to endovascular treatment and outcome in acute ischemic stroke: Mr clean registry results. Circulation, 138(3):232–240, 2018.

Christopher T Richards, Ryan Huebinger, Katie L Tataris, Joseph M We-
ber, Laura Eggers, Eddie Markul, Leslee Stein-Spencer, Kenneth S Pearlman, Jane L Holl, and Shyam Prabhakaran. Cincinnati prehospital stroke scale can identify large vessel occlusion stroke. Prehospital Emergency Care, 22(3):312–318, 2018.

Jeffrey L Saver, Mayank Goyal, AAD Van der Lugt, Bijoy K Menon, Charles BLM Majoie, Diederik W Dippel, Bruce C Campbell, Raul G
Nogueira, Andrew M Demchuk, Alejandro Tomasello, et al. Time to treat-
ment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis. Jama, 316(12):1279–1289, 2016.

Jan F Scheitz, Azmil H Abdul-Rahim, Rachael L MacIsaac, Charith Cooray, Heidi Sucharew, Dawn Kleindorfer, Pooja Khatri, Joseph P Broderick, Hein-rich J Audebert, Niaz Ahmed, et al. Clinical selection strategies to identify ischemic stroke patients with large anterior vessel occlusion: results from sits-istr (safe implementation of thrombolysis in stroke international stroke thrombolysis registry). Stroke, 48(2):290–297, 2017.

Chun-Han Wang, Yu-Chen Chang, Yung Yang, Wen-Chu Chiang, Sung-Chun
Tang, Li-Kai Tsai, Chung-Wei Lee, Jiann-Shing Jeng, Matthew Huei-Ming
Ma, Ming-Ju Hsieh, et al. Prehospital-stroke-scale parameterized hospital selection protocol for suspected stroke patients considering door-to-treatment durations. Journal of the American Heart Association, 11(7):e023760, 2022.