類別:

顱內放射手術之危及器官分割

2025-03-19T09:44:38Z

標題: 顱內放射手術之危及器官分割; Segmentation of Organs at Risk for Intracranial Radiosurgery 作者: 連薇瑛; Wei-Ying Lien 摘要: 本研究評估具自動決定超參數的深度學習框架 nnU-Net 在顱內放射手術中對危及器官自動分割的表現。基於先前腫瘤分割的研究成果，我們將研究重點擴展至危及器官的分割，以彌補現有文獻中發現的不足之處。根據最新研究進展，我們利用多模態影像開發了用於危及器官自動分割的模型。本研究使用台大醫院電腦刀中心的大規模數據集進行實驗，著重於六個重要器官：腦幹、雙側眼球、視交叉和雙側視神經。實驗方法包含三個部分：評估不同影像模態的分割準確度、探討聯合危及器官和靶體積分割的可行性，以及分析危及器官和腫瘤之間的距離關係。研究結果顯示，整合多模態的三維低解析度模型達到最佳表現。然而，視交叉因其體積小，在自動及手動描繪上都面臨重大挑戰，因此呈現相對較低的分割準確度。儘管如此，我們的危及器官模型在腦幹分割上展現了與專家手動描繪相當的準確度，並成功驗證了聯合的危及器官-目標體積分割模型的可行性，為臨床自動化應用提供了新的方向。; This study evaluates the performance of nnU-Net, a deep learning framework that automatically determines hyperparameters, in the automatic segmentation of organs at risk (OARs) for intracranial radiosurgery. Building upon previous research on tumor segmentation, we expanded our focus to include the segmentation of OARs to address the insufficiencies identified in the existing literature. Based on recent advances, we developed an automatic model for OAR segmentation utilizing multimodal imaging. The study utilizes a large-scale dataset from the CyberKnife Center at National Taiwan University Hospital, focusing on six critical organs: brainstem, bilateral eyes, optic chiasm, and bilateral optic nerves. Our experimental methodology consists of three components: evaluating segmentation accuracy across different imaging modalities, examining the feasibility of joint OAR and target volume (TV) segmentation, and analyzing their spatial relationships. The results demonstrate that the 3D low-resolution model with multimodal integration achieves optimal performance. However, the optic chiasm exhibits relatively lower segmentation accuracy, as its small volume poses significant challenges for both automatic and manual delineation. Nevertheless, our OAR model demonstrates accuracy comparable to expert manual delineation in brainstem segmentation and successfully validates the feasibility of joint OAR-TV segmentation models, offering new directions for clinical automation applications.

頸橈動脈與心電圖量測演算法系統之驗證

2024-08-15T17:39:52Z

標題: 頸橈動脈與心電圖量測演算法系統之驗證; Verification of Algorithm for Carotid and Radial artery Assessment with Electrocardiogram and Force Sensor 作者: 蔡承育; Cheng-Yu Tsai 摘要: 心腦血管阻塞所造成的疾病是常見的死亡原因，而治療後也造成嚴重的生活不便且容易再次復發。頸動脈為供給大腦血液之主要血管，若在其血管分叉處產生動脈硬化狹窄，容易因內頸動脈或外頸動脈阻塞造成缺血性腦中風與臉部麻痺失控的症狀; 在心臟方面亦有許多疾病，如：心肌缺氧、心律不整、心肌梗塞、傳導障礙…等，但由於心臟疾病沒有特異性的早期症狀，常被誤以為老化而受到忽略而錯過治療時間。頸動脈超音波為腦中風罹患篩檢的重要工具之一，但檢查方式須仰賴許多臨床人員執行，而超音波檢查也常因為有人為操作與經驗問題，其結果因環境因素與操作者而有差異。在心臟疾病方面，心電圖為評估之重要工具，但檢查需要在靜止狀態執行，於動作狀態下的心臟症狀，則不易察覺。本研究整合了不同的訊號來源而建立評估的演算法，可以分別對腦血管與心臟功能進行評估。在腦部血管檢查方面，利用非侵襲性的雙頸動脈壓力感測器所測量之訊號，透過所撰寫之峰值擷取演算法分析基礎血液動力學之參數與進階阻力指標與彈性係數，並與臨床常規頸動脈超音波檢查進行比對。而為了評估心臟方面的狀況，本研究計算心電圖訊號與橈動脈訊號的尖峰時間差距，以分析更多心臟收縮等資訊。在結果方面，經過演算法計算的結果與臨床超音波比對，在血液動力學部分，如：心跳次數、血管收縮加速度、阻力係數其相關性皆呈現高度相關，另外也可以輔助計算出心電圖與橈動脈之波形差異時間。透過壓力感測器整合峰值擷取演算法，可以自動化且有效計算出許多臨床上重要意義的血液動力學參數，透過更多的臨床試驗與數據收集分析後，將可以更有效的運用於臨床與普及檢查方式，達到輔助臨床診斷的重要功能。; Obstructions of cerebrovascular or cardiovascular are the common cause of death. Although receiving treatment, it also causes serious inconvenience in daily and it is possible to recurrent. Carotid artery is the main vessel supplying blood to brain. When the arteriosclerotic stenosis occurred in the carotid bifurcation, including internal carotid artery or external carotid artery, some symptoms were observed as ischemic stroke or facial paralysis. In addition, there are many cardiovascular diseases, such as: myocardial hypoxia, arrhythmia, myocardial infarction, conduction disorders, etc. will result in sudden cardiac death. However, lack of specific early symptoms of cardiovascular disease, often mistaken for aging and neglected to miss treatment time. Carotid ultrasound is one of the important tools for the cerebrovascular accident screening, whereas, the examination must rely on clinical professional performed. Furthermore, the results are easily affected by operator dependent and environmental factors. Besides, to evaluate cardiac disease, electrocardiogram is the essential tool for evaluating. However, procedure needed to be done at rest state that some patterns under movement are not noticeable easily. In this study, we integrated different signals and built up the system with algorithm for automatic peak detection. Hemodynamic of cerebrovascular and cardiac functions can be evaluated respectively. Noninvasive duel pressure sensor used for measuring the signals from common carotid artery. After collecting the waveform, basic information from hemodynamic and advanced parameters, including heart rate, resistance index and acceleration time were calculated. To find the correlation between this system and standard sonography, we compared the results with clinical carotid ultrasound examination for each subject. For evaluating heart functions, the peak time region between the collecting ECG signal and the radial artery signal were calculated. Based on the results, hemodynamic parameters, such as heart rate, acceleration time of vasoconstriction, and the resistance index are highly correlated with the examination by ultrasound. In addition, the peak tine region between electrocardiogram and radial artery waveform can also be calculated automatically. With more clinical trials recruited and data analyzed, it is possible to provide physician aiding clinical diagnosis.

頸椎脊髓病變患者之大腦白質完整性與感覺和運動功能相關：擴散頻譜影像研究

2021-06-17T08:40:56Z

標題: 頸椎脊髓病變患者之大腦白質完整性與感覺和運動功能相關：擴散頻譜影像研究; Cerebral white matter tract integrity is associated with sensory and motor functions in patients with cervical degenerative myelopathy: A diffusion spectrum imaging study 作者: Cheng-Hsien Peng; 彭政憲摘要: 頸痛病人日益增加，有些會發展為退化性脊椎病並導致脊髓神經病變，其臨床表徵包括感覺喪失、異常疼痛、動作失調、平衡及行走功能受限，並影響注意力、情緒及生活品質等，嚴重者須接受手術治療，但手術後有些仍有許多症狀持續，目前確實的致病機轉機制不明。先前的研究表明，脊髓型頸椎病患者在脊髓水平的平均擴散係數(MD)和分數各向異性(FA)均低於正常人。然而，這種局部變化是否會影響大腦中的白質神經束仍不清楚。在這項研究中，我們旨在比較脊髓型頸椎病患者和健康對照組之間大腦白質微觀結構的差異，並期望找出這些變化的神經束與功能指標的相關性。我們徵召了27位頸椎脊髓神經病變的病人(年齡:56.63±13.05歲，18名男性9名女性)及27位年齡性別配對之健康中老年人(年齡:55.30±13.03歲，18名男性9名女性)，所有受測者皆接受T1權重影像及擴散頻譜造影影像以探討大腦白質的完整性，包括從脊髓到大腦的上行及下行路徑，並以JOA、mJOA、NDI問卷詢問感覺動作功能。我們運用MAP-MRI的組件將擴散頻譜造影的影像計算出4種擴散指標，包括概化部分不等向性(GFA)、軸向擴散係數(AD)、平均擴散係數(MD)及徑向擴散係數(RD)。其後，運用全腦神經束自動化分析(TBAA)來獲得大腦主要76條神經束的四種擴散指標所構成的三維結構腦聯結圖(3D-connectogram)，作為我們探討白質神經纖維束結構的依據。我們使用簇群權重(TFCW)分數來進行組分析，我們計算組間每個步驟的效應量以及估計的加權分數，以找出差異最大的神經束步驟區塊。以線性迴歸分析白質完整性與感覺動作認知功能的相關性，控制年齡的因子後，探討有變化之大腦內神經束與功能之相關性。我們發現頸椎脊髓神經病變病人之脊髓神經病變造成部分上行及下行路徑的神經束病變，且此部分白質變化與動作感覺能力有相關。此外我們發現大腦內與認知功能相關的白質神經束亦有變化，此變化除了與動作感覺功能有關之外，與疼痛睡眠也有相關。顯示頸部脊髓神經束的受傷，除了對大腦有影響，對認知功能的神經束亦有影響。需再進一步探討病人的認知功能與大腦內白質變化的關係。; Introduction: Cervical myelopathy is a common degenerative condition caused by compression on the spinal cord that is characterized by clumsiness in hands and gait imbalance. Patients demonstrated multiple symptoms and signs, including sensory, motor, control and cognition relate complaints. Severe cases require surgery, but in some cases the symptoms still persist after surgery. Previous studies reported that patients with cervical myelopathy showed higher mean diffusivity (MD) and lower fractional anisotropy (FA) than normal subjects between each spinal level. However, whether this local change would affect white matter tracts in the brain is not clear. In this study, we aimed to compare the differences of cerebral white matter microstructural property between patients with cervical myelopathy and health controls, and we aimed to identify the functional correlation of the change tracts in the brain. Subjects: Two groups of participants were recruited in the study: 27 healthy older adults (age: 56.63 ± 13.05, 18 males and 9 females), 27 patients with cervical myelopathy (age: 55.30 ± 13.03, 18 males and 9 females). Imaging: All participants received T1-weighted imaging and DSI on a 3T Siemens Prisma MRI System (Siemens Medical, Erlangen, Germany) with a 32-channel phased array head coil in National Taiwan University Hospital. The parameters were as follow. T1-weighted imaging used a three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) sequence, TR/TE = 2000/3 ms, FOV = 352 x 290 x 208 mm3, flip angle = 9o, resolution = 1 x 1 x 1 mm3. Diffusion spectrum imaging (DSI) used an echo planer imaging (EPI) diffusion sequence, TR/TE = 9600/130 ms, matrix size = 80 x 80, FOV = 200 x 200 mm2, resolution = 2.5 mm, 102 diffusion encoding gradients with bmax = 4000 s/mm2. Image Quality Assurance (QA): Only images with signal-to-noise ratio (SNR) higher than 25 were included for subsequent analysis. Analysis: We used whole brain tract-based automatic analysis (TBAA) to obtain a 2D connectogram for each DSI dataset. The connectogram provides generalized fractional anisotropy (GFA), fractional anisotropy (FA), axial diffusivity (AD), mean diffusivity (MD) and radial diffusivity (RD) profiles of 76 white matter tract bundles. Each profile contained 100 sampled values at 100 equidistant steps along the tract. We used threshold free cluster weighted (TFCW) scores for group analysis. We calculated the effect size of each step between groups, and estimated weighted scores to select the most different tract steps among the two groups. We did the linear multiple regression to identify the main contributor of the tracts for the specific functional item. Result: A total of 23 segments were found to show top 5% difference in the weighted scores of GFA, AD or RD when comparing patients with cervical myelopathy with normal controls. The values of GFA of the affected segments were uniformly lower in patients. Conclusion: As expected, we found significant reduction in GFA in the sensorimotor tracts, which were supposed to be the primary affected tracts in cervical myelopathy. Moreover, we also found altered tracts that were mostly related to cognitive functions, such as the left fornix, right frontal-striatum to the ventral lateral prefrontal cortex, and the splenium of the corpus callosum. Our results are consistent with previous studies reporting that cognitive dysfunctions may be related to disorders of the cervical spine or spinal cord. We speculate that patients with cervical myelopathy may be subjected to emotion problems due to reduced mobility, which may lead to cognitive decline as reflected by the impairment of cognitive-related tracts. The relationship between cognitive function and white matter changes in the brain needs to be further studied.

預測早期思覺失調症患者之治療效果：利用大腦神經纖維束及皮質體積之機器學習模型

2021-06-17T08:31:57Z

標題: 預測早期思覺失調症患者之治療效果：利用大腦神經纖維束及皮質體積之機器學習模型; Machine learning prediction of treatment response in early schizophrenia based on cortical volume and white matter tract integrity 作者: Wei-Chia Lu; 呂維家摘要: 介紹: 對於思覺失調症患者來說，達到症狀的緩解是非常重要的，然而每位患者對於抗精神藥物的治療反應不盡相同，有些患者經過藥物的治療後卻無法達到症狀緩解的情形。而根據之前的文獻指出不同的治療效果會反映在白質及灰質微結構上(Ashburner & Friston, 2005; Bora et al., 2011; Reis Marques et al., 2014)，在這次的研究中，我們找到在早期思覺失調症中對於用藥緩解和非緩解兩組間有顯著差異的幾個部位，包含利用規範性模型計算出的z值來得到的三條神經纖維束以及利用LPBA40的模板計算出的皮質體積，我們將得到的三條神經纖維束以及皮質體積並藉由機器學習的模型來達到個別化預測早期思覺失調症藥物治療後的症狀反應情形。材料及方法: 本次實驗納入65位的思覺失調症患者(病程皆小於三年內，29位男性，36位女性，年齡:26.9±6.2年)，和50位健康的受試者，後分為52位於訓練組(年齡:26.4±6.1年，26位男性，26位女性)以及13位於測試組(年齡:28.7±6.6年，男性3位，女性10位)。所有的受試者皆在台灣大學醫院的3T西門子Tim Trio磁振造影系統以32通道線圈來掃描獲得T1加權影像以及擴散頻譜影像。影像品質確保:僅有影像的訊號雜訊比大於20以及T1-綜合非等項性指標之相關係數大於0.5的個案會被納入此研究分析:我們利用散頻譜影像與全腦白質纖維神經束分析方式(Chen et al., 2015)來獲得大腦主要76條白質神經束的完整度，並利用779位其年齡涵蓋5至80歲的正常族群資料，依據性別分組，並計算出患者的標準偏差值，經由統計分析後得到三條神經束在緩解與非緩解組中有相當程度的差異，此三條神經纖維為右側弓狀束、右側下枕額束及右側鉤束。此外由CAT12中的LPBA40模板計算出皮質體積的資訊，並得到在緩解與非緩解組中尾狀核有顯著性的差異。患者被分類為緩解及非緩解是藉由活性與負性症狀量表、思覺失調症緩解工作小組(Nancy C. Andreasen et al., 2005)以及台灣大學醫院的精神科醫師來診斷分類，最後藉由MATLAB中的classification learner來得到最有潛力的機器學習模型以預測治療的反應。結果: 65位思覺失調症患者皆完成至少6個月的追蹤評定，根據思覺失調症緩解工作小組以及台灣大學醫院的精神科醫師診斷，52位訓練組中有37位為症狀緩解，15位為非緩解; 13位測試組中有9位為症狀緩解，4位為非緩解。而在統計分析後得到在緩解與非緩解中有顯著性差異的三條神經束及腦區，利用機器學習的方式來得到最有潛力分辨緩解與非緩解的模型，在訓練組中的正確分辨率為: Decision Tree: 0.923, RUSBoosted Trees: 0.827, Fine KNN: 0.808，套用到訓練組中得到個案的預測準確率為Decision Tree: 0.846, RUSBoosted Trees: 0.615, Fine KNN: 0.615. 結論: 我們的目標是找到對於早期思覺失調症治療反應的生物標記並且將組間比較邁向個體化預測，而在這次實驗中我們得到藉由決策樹訓練出來的機器學習模型在於個案的預測準確率有84.6%，是非常有潛力來達到個體化預測的模型。對於思覺失調症而言，一個長期的研究計畫是非常重要且有價值的，我們可以在此實驗中得到神經束以及皮質體積有其貢獻在預測治療反應。並冀望結合更多臨床的評估及其他神經影像資訊，此模型將可作為醫師擬定個案治療計畫之參考。; Introduction: It is important to achieve symptomatic remission for the treatment of schizophrenia. However, some patients don't achieve remission state even after antipsychotics treatment. Different treatment outcomes may be reflected in the white matter microstructure (Ashburner & Friston, 2005; Bora et al., 2011; Reis Marques et al., 2014). In this study, we identified three white mater tracts calculate by adjusted Z-scores of diffusion indices and cortical volume by LPBA40 atlas that showed significant difference between remitted and non-remitted patients in early schizophrenia. We combined those three bundles and cortical volume with machine learning to promote prediction accuracy for treatment outcomes in early schizophrenia. Materials and Methods: sixty-five patients with schizophrenia (course of disease < 3 years, age: 26.864 ± 6.224, 29 males and 36 females) and 50 healthy control were recruited. Separate into training group (age: 26.404 ± 6.026, 26 males and 26 females) and testing group (age: 28.705 ± 6.65, 3 males and 10 females). Imaging: All participants received T1-weighted imaging and DSI on a 3T Siemens Tim Trio MRI System with 32-channel phased array head coil in National Taiwan University Hospital. Image Quality Assurance: Only images with signal-to-noise ratio higher than 20 and T1-GFA correlation coefficient higher than 0.5 were included for subsequent analysis. Analysis: We used TBAA to obtain profiles of diffusion indices (Chen et al., 2015), and calculated GFA, AD, RD, MD adjusted Z-scores made by 779 healthy subjects, corresponding to three tracts of interest. These three tracts included the right arcuate fasciculus, the right inferior frontal occipital fasciculus, and the right uncinate fasciculus. Furthermore, calculate cortical volume using LPBA40 atlas by CAT12. Right caudate have significant difference. Patients were grouped into good or poor treatment outcomes according to the PANSS scores, RSWG criteria (Nancy C. Andreasen et al., 2005) and clinical diagnosis by National Taiwan University Hospital. We used the classification learner from MATLAB to determine the most effective model, which had great potential for predicting treatment outcomes. Result: 65 patients had completed the 6-month follow-up assessment. According to the RSWG criteria and clinical diagnosis, 37 patients were assigned to the remission group and 15 patients were assigned to the non-remission group inside the training group and 9 and 4 patients were assigned to remission and non-remission group inside the testing group. According to the statistics, three white matter tracts and the cortical volume of right caudate show significant difference between remitted and non-remitted patients in early schizophrenia. Based on these three tract bundles and cortical volume, we used machine learning to calculate those feature together to determine the most potential model to predict treatment outcomes. The accuracy of the prediction model in training group: Decision Tree: 0.923, RUSBoosted Trees: 0.827, Fine KNN: 0.808. With the testing data, the accuracy of each model as follow: Decision Tree: 0.846, RUSBoosted Trees: 0.615, Fine KNN: 0.615. Conclusion: Our goal is to find biomarkers of treatment response in early schizophrenia. We approach the goal from group comparison to individualized prediction. Model training by decision tree is potentially useful in treatment response prediction for each individual patient. A longitudinal study on early state schizophrenia is required to validate the clinical value of these tracts and cortical volume in predicting treatment response.