Xiang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xiang Liu has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Bin Yi
Department of Anesthesiology, Southwest Hospital, Army Medical University, Chongqing, China
2 shared publications - Yiziting Zhu
Department of Anesthesiology, Southwest Hospital, Third Military Medical University, Chongqing, China
2 shared publications - Hongqian Wang
Shanghai Jiao Tong University,Department of Micro/Nano Electrorics,Shanghai,China
1 shared publication - Qilong Sun
Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China
1 shared publication - Xiao-Yan Hu
Department of Anaesthesiology, Southwest Hospital, Third Military Medical University
1 shared publication - Xiaoguang Lin
Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing, China
1 shared publication
Latest Publications
- Open Access
TSP-OCS: A Time-Series Prediction for Optimal Camera Selection in Multi-Viewpoint Surgical Video Analysis
Recording the open surgery process is essential for educational and medical evaluation purposes; however, traditional single-camera methods often face challenges such as occlusions caused by the surgeon's head and body, as well as limitations due to fixed camera angles, which reduce comprehensibility of the video content. This study addresses these limitations by employing a multi-viewpoint camera recording system, capturing the surgical procedure from six different angles to mitigate occlusions. We propose a fully supervised learning-based time series prediction method to choose the best shot sequences from multiple simultaneously recorded video streams, ensuring optimal viewpoints at each moment. Our time series prediction model forecasts future camera selections by extracting and fusing visual and semantic features from surgical videos using pre-trained models. These features are processed by a temporal prediction network with TimeBlocks to capture sequential dependencies. A linear embedding layer reduces dimensionality, and a Softmax classifier selects the optimal camera view based on the highest probability. In our experiments, we created five groups of open thyroidectomy videos, each with simultaneous recordings from six different angles. The results demonstrate that our method achieves competitive accuracy compared to traditional supervised methods, even when predicting over longer time horizons. Furthermore, our approach outperforms state-of-the-art time series prediction techniques on our dataset. This manuscript makes a unique contribution by presenting an innovative framework that advances surgical video analysis techniques, with significant implications for improving surgical education and patient safety.
Open AccessTSP-OCS: A Time-Series Prediction for Optimal Camera Selection in Multi-Viewpoint Surgical Video Analysis
Recording the open surgery process is essential for educational and medical evaluation purposes; however, traditional single-camera methods often face challenges such as occlusions caused by the surgeon's head and body, as well as limitations due to fixed camera angles, which reduce comprehensibility of the video content. This study addresses these limitations by employing a multi-viewpoint camera recording system, capturing the surgical procedure from six different angles to mitigate occlusions. We propose a fully supervised learning-based time series prediction method to choose the best shot sequences from multiple simultaneously recorded video streams, ensuring optimal viewpoints at each moment. Our time series prediction model forecasts future camera selections by extracting and fusing visual and semantic features from surgical videos using pre-trained models. These features are processed by a temporal prediction network with TimeBlocks to capture sequential dependencies. A linear embedding layer reduces dimensionality, and a Softmax classifier selects the optimal camera view based on the highest probability. In our experiments, we created five groups of open thyroidectomy videos, each with simultaneous recordings from six different angles. The results demonstrate that our method achieves competitive accuracy compared to traditional supervised methods, even when predicting over longer time horizons. Furthermore, our approach outperforms state-of-the-art time series prediction techniques on our dataset. This manuscript makes a unique contribution by presenting an innovative framework that advances surgical video analysis techniques, with significant implications for improving surgical education and patient safety.
- Open Access
- 26 August 2024
- Published by Springer Nature in Anesthesiology and Perioperative Science
- Vol. 2 (3) , 1-11
The applications and prospects of big data in perioperative anesthetic management
Perioperative anesthetic management entails a multitude of decision-making processes within complex medical scenarios. These demand the continuous and dynamic execution of precise decisions which poses significant challenges. In the age of big data, the exponential growth in data volume from diverse sources has revolutionized many fields, including healthcare, finance, and marketing. Machine learning has emerged as a powerful tool for analyzing big data, enabling the handling of large datasets and uncovering intricate patterns and relationships. The application of big data and artificial intelligence algorithms is gradually being integrated, enabling effective task completion in various stages of perioperative management, including risk prediction, decision support, and auxiliary examination. Through in-depth analysis of big data, healthcare professionals can gain insights into patient prognoses. This review provides a comprehensive overview of the distinctive features of perioperative big data and its applications in anesthesia management during the perioperative period.
- 26 August 2024
- Published by Springer Nature in Anesthesiology and Perioperative Science
- Vol. 2 (3) , 1-11
Open AccessThe applications and prospects of big data in perioperative anesthetic management
Perioperative anesthetic management entails a multitude of decision-making processes within complex medical scenarios. These demand the continuous and dynamic execution of precise decisions which poses significant challenges. In the age of big data, the exponential growth in data volume from diverse sources has revolutionized many fields, including healthcare, finance, and marketing. Machine learning has emerged as a powerful tool for analyzing big data, enabling the handling of large datasets and uncovering intricate patterns and relationships. The application of big data and artificial intelligence algorithms is gradually being integrated, enabling effective task completion in various stages of perioperative management, including risk prediction, decision support, and auxiliary examination. Through in-depth analysis of big data, healthcare professionals can gain insights into patient prognoses. This review provides a comprehensive overview of the distinctive features of perioperative big data and its applications in anesthesia management during the perioperative period.
- Open Access
- 4 April 2024
- Published by Springer Nature in Research Square
Automatic Non-Invasive Prediction of Hemoglobin Using a Deep Learning-Assisted Smartphone-Based System
Background The context and purpose of this study is to introduce a novel, compact, and efficient system that leverages deep learning and smartphone technology to estimate hemoglobin levels. Accurate measurement of hemoglobin concentration is essential for various medical scenarios, including preoperative evaluations and determining blood loss. Current models, due to their complex parameters, are not well-suited for mobile medical settings, which limits the ability to conduct frequent and rapid testing. Methods The study employed a smartphone application to capture images of the eye, which were then analyzed by a deep neural network trained using invasive blood test data. For the task of eyelid segmentation, the EGE-Unet model was used. The performance of this model was evaluated using statistical metrics including mean intersection over union (MIOU), F1 Score, accuracy, specificity, and sensitivity. For hemoglobin level prediction, the DHA(C3AE) model was employed. The performance of this model was assessed using mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), and R^2. Results The EGE-Unet model demonstrated robust performance in eyelid segmentation, achieving a MIOU of 0.78, an F1 Score of 0.87, an accuracy of 0.97, a specificity of 0.98, and a sensitivity of 0.86. The DHA(C3AE) model for hemoglobin level prediction yielded promising outcomes with a MAE of 1.34, a MAPE of 2.85, an RMSE of 1.69, and an R^2 of 0.34. The overall size of the model is modest at 1.08M, with a computational complexity of 0.12 FLOPs (G). Conclusions This groundbreaking approach eliminates the necessity for supplementary devices, providing a cost-effective, swift, and accurate method for healthcare professionals to enhance treatment planning and improve patient care in perioperative environments. The proposed system has the potential to enable frequent and rapid testing of hemoglobin levels, which can be particularly beneficial in mobile medical settings. Trial Registration: The clinical trial was registered on the Chinese Clinical Trial Registry (No. ChiCTR2100044138) on March 11, 2021.
- 4 April 2024
- Published by Springer Nature in Research Square
Open AccessAutomatic Non-Invasive Prediction of Hemoglobin Using a Deep Learning-Assisted Smartphone-Based System
Background The context and purpose of this study is to introduce a novel, compact, and efficient system that leverages deep learning and smartphone technology to estimate hemoglobin levels. Accurate measurement of hemoglobin concentration is essential for various medical scenarios, including preoperative evaluations and determining blood loss. Current models, due to their complex parameters, are not well-suited for mobile medical settings, which limits the ability to conduct frequent and rapid testing. Methods The study employed a smartphone application to capture images of the eye, which were then analyzed by a deep neural network trained using invasive blood test data. For the task of eyelid segmentation, the EGE-Unet model was used. The performance of this model was evaluated using statistical metrics including mean intersection over union (MIOU), F1 Score, accuracy, specificity, and sensitivity. For hemoglobin level prediction, the DHA(C3AE) model was employed. The performance of this model was assessed using mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), and R^2. Results The EGE-Unet model demonstrated robust performance in eyelid segmentation, achieving a MIOU of 0.78, an F1 Score of 0.87, an accuracy of 0.97, a specificity of 0.98, and a sensitivity of 0.86. The DHA(C3AE) model for hemoglobin level prediction yielded promising outcomes with a MAE of 1.34, a MAPE of 2.85, an RMSE of 1.69, and an R^2 of 0.34. The overall size of the model is modest at 1.08M, with a computational complexity of 0.12 FLOPs (G). Conclusions This groundbreaking approach eliminates the necessity for supplementary devices, providing a cost-effective, swift, and accurate method for healthcare professionals to enhance treatment planning and improve patient care in perioperative environments. The proposed system has the potential to enable frequent and rapid testing of hemoglobin levels, which can be particularly beneficial in mobile medical settings. Trial Registration: The clinical trial was registered on the Chinese Clinical Trial Registry (No. ChiCTR2100044138) on March 11, 2021.