Yueh-Hsun Tsai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yueh-Hsun Tsai has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Tsai-Wang Huang
Division of Thoracic Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan
3 shared publications - Cheng-Jung Lin
Department of Thoracic Surgery, School of Medicine, College of Medicine, National Defense Medical University, Taipei, Taiwan
2 shared publications - Hsiang Teng
Division of Thoracic Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan
2 shared publications - Hsu-Kai Huang
Division of Thoracic Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan
2 shared publications - Kuan-Hsun Lin
Division of Thoracic Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan
2 shared publications - Ying-Shian Chen
Department of Thoracic Surgery, School of Medicine, College of Medicine, National Defense Medical University, Taipei, Taiwan
2 shared publications
Latest Publications
- Open Access
- 9 September 2026
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 10, e94084
Real-Time AI-Augmented Fluoroscopic Navigation for Intraoperative Pulmonary Nodule Localization: Prospective Observational Pilot Study
Background: The integration of AI into intraoperative surgical imaging represents an emerging frontier in digital health. Despite advances in preoperative computed tomography (CT)–based surgical planning, real-time translation of imaging data into actionable intraoperative guidance remains limited by CT-to-body divergence—a fundamental information gap between preoperative digital models and the dynamic surgical field. This divergence, driven by lung deflation under anesthesia and positional changes, represents a critical digital-to-physical registration challenge that current preoperative imaging workflows fail to address in real time. Objective: This study aimed to evaluate the clinical feasibility, localization success, and safety of the LungVision system—an AI-augmented fluoroscopic navigation platform—for real-time intraoperative localization of small pulmonary nodules during thoracoscopic surgery. Methods: A prospective single-center study enrolled 14 patients with pulmonary nodules requiring localization prior to thoracoscopic resection between January 2024 and December 2024. The platform comprises a passive radiopaque positioning board, an AI-powered computing unit for real-time image processing, and a tablet-based interface for procedural planning and augmented visualization. All patients received dual localization with either preoperative CT-guided dye injection or Archimedes virtual bronchoscopic navigation followed by intraoperative localization with the LungVision system and video-assisted thoracoscopic surgery. Demographic data, lesion characteristics, procedural performance, and procedure-related complications were recorded. Results: The mean patient age was 57.2 (SD 9.2) years, and 92.9% (13/14) were nonsmokers. Most nodules were peripherally located (12/14, 85.7%), with a mean diameter of 9.3 (SD 5.3) mm and a mean CT attenuation of −320.1 (SD 334.9) Hounsfield units. LungVision successfully localized all target lesions intraoperatively, with a mean navigation time of 38.6 (SD 19.5) minutes. Complete resection was achieved in all cases, and 71.4% (10/14) of nodules were pathologically malignant. No intraoperative or localization-related complications were observed. The system was integrated into the existing operating room without additional infrastructure modifications. Conclusions: In this prospective study, the LungVision system achieved successful intraoperative localization of small, hypodense pulmonary nodules using a bronchoscopic approach integrated with conventional C-arm fluoroscopy. These findings support the feasibility of the technique and provide preliminary evidence on its use in thoracoscopic resection workflows. Larger studies are needed to further evaluate its clinical performance and implementation in diverse practice settings. Trial Registration: ClinicalTrials.gov NCT07682012; https://clinicaltrials.gov/study/NCT07682012
- 9 September 2026
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 10, e94084
Open AccessReal-Time AI-Augmented Fluoroscopic Navigation for Intraoperative Pulmonary Nodule Localization: Prospective Observational Pilot Study
Background: The integration of AI into intraoperative surgical imaging represents an emerging frontier in digital health. Despite advances in preoperative computed tomography (CT)–based surgical planning, real-time translation of imaging data into actionable intraoperative guidance remains limited by CT-to-body divergence—a fundamental information gap between preoperative digital models and the dynamic surgical field. This divergence, driven by lung deflation under anesthesia and positional changes, represents a critical digital-to-physical registration challenge that current preoperative imaging workflows fail to address in real time. Objective: This study aimed to evaluate the clinical feasibility, localization success, and safety of the LungVision system—an AI-augmented fluoroscopic navigation platform—for real-time intraoperative localization of small pulmonary nodules during thoracoscopic surgery. Methods: A prospective single-center study enrolled 14 patients with pulmonary nodules requiring localization prior to thoracoscopic resection between January 2024 and December 2024. The platform comprises a passive radiopaque positioning board, an AI-powered computing unit for real-time image processing, and a tablet-based interface for procedural planning and augmented visualization. All patients received dual localization with either preoperative CT-guided dye injection or Archimedes virtual bronchoscopic navigation followed by intraoperative localization with the LungVision system and video-assisted thoracoscopic surgery. Demographic data, lesion characteristics, procedural performance, and procedure-related complications were recorded. Results: The mean patient age was 57.2 (SD 9.2) years, and 92.9% (13/14) were nonsmokers. Most nodules were peripherally located (12/14, 85.7%), with a mean diameter of 9.3 (SD 5.3) mm and a mean CT attenuation of −320.1 (SD 334.9) Hounsfield units. LungVision successfully localized all target lesions intraoperatively, with a mean navigation time of 38.6 (SD 19.5) minutes. Complete resection was achieved in all cases, and 71.4% (10/14) of nodules were pathologically malignant. No intraoperative or localization-related complications were observed. The system was integrated into the existing operating room without additional infrastructure modifications. Conclusions: In this prospective study, the LungVision system achieved successful intraoperative localization of small, hypodense pulmonary nodules using a bronchoscopic approach integrated with conventional C-arm fluoroscopy. These findings support the feasibility of the technique and provide preliminary evidence on its use in thoracoscopic resection workflows. Larger studies are needed to further evaluate its clinical performance and implementation in diverse practice settings. Trial Registration: ClinicalTrials.gov NCT07682012; https://clinicaltrials.gov/study/NCT07682012
- Open Access
- 25 February 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Real-Time AI-Augmented Fluoroscopic Navigation for Intraoperative Pulmonary Nodule Localization: Prospective Observational Pilot Study (Preprint)
BACKGROUND The integration of AI into intraoperative surgical imaging represents an emerging frontier in digital health. Despite advances in preoperative computed tomography (CT)–based surgical planning, real-time translation of imaging data into actionable intraoperative guidance remains limited by CT-to-body divergence—a fundamental information gap between preoperative digital models and the dynamic surgical field. This divergence, driven by lung deflation under anesthesia and positional changes, represents a critical digital-to-physical registration challenge that current preoperative imaging workflows fail to address in real time. OBJECTIVE This study aimed to evaluate the clinical feasibility, localization success, and safety of the LungVision system—an AI-augmented fluoroscopic navigation platform—for real-time intraoperative localization of small pulmonary nodules during thoracoscopic surgery. METHODS A prospective single-center study enrolled 14 patients with pulmonary nodules requiring localization prior to thoracoscopic resection between January 2024 and December 2024. The platform comprises a passive radiopaque positioning board, an AI-powered computing unit for real-time image processing, and a tablet-based interface for procedural planning and augmented visualization. All patients received dual localization with either preoperative CT-guided dye injection or Archimedes virtual bronchoscopic navigation followed by intraoperative localization with the LungVision system and video-assisted thoracoscopic surgery. Demographic data, lesion characteristics, procedural performance, and procedure-related complications were recorded. RESULTS The mean patient age was 57.2 (SD 9.2) years, and 92.9% (13/14) were nonsmokers. Most nodules were peripherally located (12/14, 85.7%), with a mean diameter of 9.3 (SD 5.3) mm and a mean CT attenuation of −320.1 (SD 334.9) Hounsfield units. LungVision successfully localized all target lesions intraoperatively, with a mean navigation time of 38.6 (SD 19.5) minutes. Complete resection was achieved in all cases, and 71.4% (10/14) of nodules were pathologically malignant. No intraoperative or localization-related complications were observed. The system was integrated into the existing operating room without additional infrastructure modifications. CONCLUSIONS In this prospective study, the LungVision system achieved successful intraoperative localization of small, hypodense pulmonary nodules using a bronchoscopic approach integrated with conventional C-arm fluoroscopy. These findings support the feasibility of the technique and provide preliminary evidence on its use in thoracoscopic resection workflows. Larger studies are needed to further evaluate its clinical performance and implementation in diverse practice settings. CLINICALTRIAL ClinicalTrials.gov NCT07682012; https://clinicaltrials.gov/study/NCT07682012
- 25 February 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessReal-Time AI-Augmented Fluoroscopic Navigation for Intraoperative Pulmonary Nodule Localization: Prospective Observational Pilot Study (Preprint)
BACKGROUND The integration of AI into intraoperative surgical imaging represents an emerging frontier in digital health. Despite advances in preoperative computed tomography (CT)–based surgical planning, real-time translation of imaging data into actionable intraoperative guidance remains limited by CT-to-body divergence—a fundamental information gap between preoperative digital models and the dynamic surgical field. This divergence, driven by lung deflation under anesthesia and positional changes, represents a critical digital-to-physical registration challenge that current preoperative imaging workflows fail to address in real time. OBJECTIVE This study aimed to evaluate the clinical feasibility, localization success, and safety of the LungVision system—an AI-augmented fluoroscopic navigation platform—for real-time intraoperative localization of small pulmonary nodules during thoracoscopic surgery. METHODS A prospective single-center study enrolled 14 patients with pulmonary nodules requiring localization prior to thoracoscopic resection between January 2024 and December 2024. The platform comprises a passive radiopaque positioning board, an AI-powered computing unit for real-time image processing, and a tablet-based interface for procedural planning and augmented visualization. All patients received dual localization with either preoperative CT-guided dye injection or Archimedes virtual bronchoscopic navigation followed by intraoperative localization with the LungVision system and video-assisted thoracoscopic surgery. Demographic data, lesion characteristics, procedural performance, and procedure-related complications were recorded. RESULTS The mean patient age was 57.2 (SD 9.2) years, and 92.9% (13/14) were nonsmokers. Most nodules were peripherally located (12/14, 85.7%), with a mean diameter of 9.3 (SD 5.3) mm and a mean CT attenuation of −320.1 (SD 334.9) Hounsfield units. LungVision successfully localized all target lesions intraoperatively, with a mean navigation time of 38.6 (SD 19.5) minutes. Complete resection was achieved in all cases, and 71.4% (10/14) of nodules were pathologically malignant. No intraoperative or localization-related complications were observed. The system was integrated into the existing operating room without additional infrastructure modifications. CONCLUSIONS In this prospective study, the LungVision system achieved successful intraoperative localization of small, hypodense pulmonary nodules using a bronchoscopic approach integrated with conventional C-arm fluoroscopy. These findings support the feasibility of the technique and provide preliminary evidence on its use in thoracoscopic resection workflows. Larger studies are needed to further evaluate its clinical performance and implementation in diverse practice settings. CLINICALTRIAL ClinicalTrials.gov NCT07682012; https://clinicaltrials.gov/study/NCT07682012
- Open Access
Intrapericardial Thymoma Presented as Pericardial Tamponade with Post-Operative Myasthenia Gravis
Background: Thymoma is an epithelial mass arising from the thymus. Most thymomas are located in the anterior mediastinum. Ectopic intrapericardial thymoma is very unusual; to date, only eight cases of pericardial thymoma have been reported. Among thymoma patients, 20% to 25% are associated with myasthenia gravis. However, postoperative myasthenia gravis occurs in less than 1% of cases. Here, we share a rare case of ectopic intrapericardial thymoma that developed postoperative myasthenia gravis six months after surgery. Case presentation: A 66-year-old woman visited the outpatient department due to productive cough and chest pain. Chest radiography showed increased soft tissue opacity over the mediastinum. A soft tissue mass in the pericardium and a ground glass nodule in right upper lung were noted using chest computed tomography. The diagnosis of thymoma, type B2, pT3N0M0, and stage IIIA and synchronous adenocarcinoma in situ of the right upper lung was confirmed after surgical removal. Six months later, the patient developed postoperative myasthenia gravis. Conclusions: Thymoma is rarely considered a differential diagnosis in pericardial tumors. Surgical removal with adjuvant radiation therapy should be performed considering the malignancy potential of thymomas and cardiac complications. In patients without myasthenia gravis, a small chance of postoperative myasthenia gravis remains. Patients should be carefully monitored for myasthenia gravis after surgery.
Open AccessIntrapericardial Thymoma Presented as Pericardial Tamponade with Post-Operative Myasthenia Gravis
Background: Thymoma is an epithelial mass arising from the thymus. Most thymomas are located in the anterior mediastinum. Ectopic intrapericardial thymoma is very unusual; to date, only eight cases of pericardial thymoma have been reported. Among thymoma patients, 20% to 25% are associated with myasthenia gravis. However, postoperative myasthenia gravis occurs in less than 1% of cases. Here, we share a rare case of ectopic intrapericardial thymoma that developed postoperative myasthenia gravis six months after surgery. Case presentation: A 66-year-old woman visited the outpatient department due to productive cough and chest pain. Chest radiography showed increased soft tissue opacity over the mediastinum. A soft tissue mass in the pericardium and a ground glass nodule in right upper lung were noted using chest computed tomography. The diagnosis of thymoma, type B2, pT3N0M0, and stage IIIA and synchronous adenocarcinoma in situ of the right upper lung was confirmed after surgical removal. Six months later, the patient developed postoperative myasthenia gravis. Conclusions: Thymoma is rarely considered a differential diagnosis in pericardial tumors. Surgical removal with adjuvant radiation therapy should be performed considering the malignancy potential of thymomas and cardiac complications. In patients without myasthenia gravis, a small chance of postoperative myasthenia gravis remains. Patients should be carefully monitored for myasthenia gravis after surgery.
- Open Access
The Role of Power Doppler Ultrasonography in Caudal Epidural Injection
Background and Objectives: Although the ultrasound-guided technique is used in caudal epidural injections, severe complications can happen if ultrasound cannot identify the occurrence of intravascular injection. To determine intraepidural and intravascular injection during caudal epidural injections, we used power Doppler ultrasonography (PDU) when injecting medications into the epidural space. Materials and Methods: This is a retrospective study that enrolled a total of 277 patients with refractory low back pain or degenerative disc from January 2019 to December 2019. The injectate flow of caudal epidural injections was examined with the assistance of PDU and confirmed by fluoroscopy. Four flow patterns were identified by PDU in our study: the “Earthworm sign,” the “Patch sign,” the ”Tubular sign” and the “Absent flow sign.” The accuracy of PDU in identifying intraepidural and intravascular injections was determined by fluoroscopy images recorded during each injection. Results: We evaluated 277 patients (mean age, 68.6 ± 13.2 years; 106 men). The “Patch sign” showed a sensitivity of 88.76% and a specificity of 80% in predicting epidural injection without intravascular injection. The “Earthworm sign” demonstrated a sensitivity of 70% and a specificity of 100% in detecting intravascular injection. The “Tubular sign” showed a specificity of 100% and a sensitivity of 9.4% in predicting successful epidural injection. The absence of a flow signal showed a sensitivity of 1.87% and a specificity of 90% in predicting successful epidural injection. Conclusions: Ultrasound-guided caudal epidural injection can accurately determine intraepidural and intravascular injections with the assistance of PDU and is thus a good alternative technique to fluoroscopy-guided caudal epidural injection.
Open AccessThe Role of Power Doppler Ultrasonography in Caudal Epidural Injection
Background and Objectives: Although the ultrasound-guided technique is used in caudal epidural injections, severe complications can happen if ultrasound cannot identify the occurrence of intravascular injection. To determine intraepidural and intravascular injection during caudal epidural injections, we used power Doppler ultrasonography (PDU) when injecting medications into the epidural space. Materials and Methods: This is a retrospective study that enrolled a total of 277 patients with refractory low back pain or degenerative disc from January 2019 to December 2019. The injectate flow of caudal epidural injections was examined with the assistance of PDU and confirmed by fluoroscopy. Four flow patterns were identified by PDU in our study: the “Earthworm sign,” the “Patch sign,” the ”Tubular sign” and the “Absent flow sign.” The accuracy of PDU in identifying intraepidural and intravascular injections was determined by fluoroscopy images recorded during each injection. Results: We evaluated 277 patients (mean age, 68.6 ± 13.2 years; 106 men). The “Patch sign” showed a sensitivity of 88.76% and a specificity of 80% in predicting epidural injection without intravascular injection. The “Earthworm sign” demonstrated a sensitivity of 70% and a specificity of 100% in detecting intravascular injection. The “Tubular sign” showed a specificity of 100% and a sensitivity of 9.4% in predicting successful epidural injection. The absence of a flow signal showed a sensitivity of 1.87% and a specificity of 90% in predicting successful epidural injection. Conclusions: Ultrasound-guided caudal epidural injection can accurately determine intraepidural and intravascular injections with the assistance of PDU and is thus a good alternative technique to fluoroscopy-guided caudal epidural injection.
- Open Access
Post-Cholecystectomy Right Hepatic Artery Pseudoaneurysm Induced by Generalized Seizure: A Case Report
Pseudoaneurysm is a rare complication of laparoscopic cholecystectomy (LC). In most cases, the patient presents with gastrointestinal bleeding or hemoperitoneum. Here, we present a case with a post-cholecystectomy right hepatic artery pseudoaneurysm (PSA) induced by a generalized seizure. A 39-year-old male was sent to the emergency room with a generalized seizure and a loss of consciousness for approximately 5 min. Diffuse abdominal pain was complained of after consciousness returned. The surgical history of LC 13 days prior was mentioned. Abdominal computer tomography (CT) revealed a lobulated fluid accumulation in the gallbladder fossa with prominent fatty stranding and suspected biloma formation. After admission for one week, sharp abdominal pain was observed. Abdominal CT angiography revealed a right hepatic artery pseudoaneurysm. Transcatheter arterial embolization was performed with a total of seven platinum coils. In conclusion, it is important for doctors to take pseudoaneurysm into consideration in the patient who presents with seizure attack after receiving LC. Late discovery of PSA when it is ruptured can lead to fatal conditions, such as severe hemoperitoneum.
Open AccessPost-Cholecystectomy Right Hepatic Artery Pseudoaneurysm Induced by Generalized Seizure: A Case Report
Pseudoaneurysm is a rare complication of laparoscopic cholecystectomy (LC). In most cases, the patient presents with gastrointestinal bleeding or hemoperitoneum. Here, we present a case with a post-cholecystectomy right hepatic artery pseudoaneurysm (PSA) induced by a generalized seizure. A 39-year-old male was sent to the emergency room with a generalized seizure and a loss of consciousness for approximately 5 min. Diffuse abdominal pain was complained of after consciousness returned. The surgical history of LC 13 days prior was mentioned. Abdominal computer tomography (CT) revealed a lobulated fluid accumulation in the gallbladder fossa with prominent fatty stranding and suspected biloma formation. After admission for one week, sharp abdominal pain was observed. Abdominal CT angiography revealed a right hepatic artery pseudoaneurysm. Transcatheter arterial embolization was performed with a total of seven platinum coils. In conclusion, it is important for doctors to take pseudoaneurysm into consideration in the patient who presents with seizure attack after receiving LC. Late discovery of PSA when it is ruptured can lead to fatal conditions, such as severe hemoperitoneum.