PURPOSE: Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive w...
详细信息
PURPOSE: Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive warnings and semi-autonomous robotic assistance or improve training of surgeons via data-driven feedback. In surgical workflow analysis up to 91% average precision has been reported for phase recognition on an open data single-center video dataset. In this work we investigated the generalizability of phase recognition algorithms in a multi-center setting including more difficult recognition tasks such as surgical action and surgical skill. METHODS: To achieve this goal, a dataset with 33 laparoscopic cholecystectomy videos from three surgical centers with a total operation time of 22 hours was created. Labels included framewise annotation of seven surgical phases with 250 phase transitions, 5514 occurences of four surgical actions, 6980 occurences of 21 surgical instruments from seven instrument categories and 495 skill classifications in five skill dimensions. The dataset was used in the 2019 international Endoscopic Vision challenge, sub-challenge for surgical workflow and skill analysis. Here, 12 research teams trained and submitted their machine learning algorithms for recognition of phase, action, instrument and/or skill assessment. RESULTS: F1-scores were achieved for phase recognition between 23.9% and 67.7% (n=9 teams), for instrument presence detection between 38.5% and 63.8% (n=8 teams), but for action recognition only between 21.8% and 23.3% (n=5 teams). The average absolute error for skill assessment was 0.78 (n=1 team). CONCLUSION: Surgical workflow and skill analysis are promising technologies to support the surgical team, but are not solved yet, as shown by our comparison of machine learning algorithms. This novel HeiChole benchmark can be used for comparable evaluation and validation of future work. In future studies, it is of utmost impo
The need for algorithms able to solve Reinforcement learning (RL) problems with few trials has motivated the advent of model-based RL methods. The reported performance of model-based algorithms has dramatically increa...
详细信息
Medical imaging has been employed to support medical diagnosis and treatment. It may also provide crucial information to surgeons to facilitate optimal surgical preplanning and perioperative management. Essentially, s...
详细信息
This work studies the deep learning-based numerical algorithms for optimal hedging problems in markets with general convex transaction costs on the trading rates, focusing on their scalability of trading time horizon....
详细信息
The need to evaluate treatment effectiveness is ubiquitous in most of empirical science, and interest in flexibly investigating effect heterogeneity is growing rapidly. To do so, a multitude of model-agnostic, nonpara...
详细信息
Reward learning is a fundamental problem in human-robot interaction to have robots that operate in alignment with what their human user wants. Many preference-based learning algorithms and active querying techniques h...
详细信息
Using recent machine learning results that present an information-theoretic perspective on underfitting and overfitting, we prove that deciding whether an encodable learning algorithm will always underfit a dataset, e...
详细信息
In this paper, we study the generalization properties of Model-Agnostic Meta-learning (MAML) algorithms for supervised learning problems. We focus on the setting in which we train the MAML model over m tasks, each wit...
详细信息
In this work, we design provably (more) efficient imitation learning algorithms that directly optimize policies from expert demonstrations. Firstly, when the transition function is known, we build on the nearly minima...
详细信息
Currently, we have witnessed the rapid development of data-driven machine learning methods, which have achieved very effective results in communication systems. Kernel learning is a typical nonlinear learning method i...
详细信息
Currently, we have witnessed the rapid development of data-driven machine learning methods, which have achieved very effective results in communication systems. Kernel learning is a typical nonlinear learning method in the machine learning community. This article proposes two novel correntropy-based kernel learning algorithms to improve the accuracy of indoor positioning in WiFi-based wireless networks. In general, correntropy as a measure of local similarity defined in kernel space can be used for robust signal processing to address large outliers. Then, through the combination of the maximum mixture correntropy criterion (MMCC) and online vector quantization (VQ), we develop a learning algorithm, named quantized kernel MMCC (QKMMCC) method, which works with the advantage of correntropy while effectively suppressing the growth of memory structure and reducing the computation in this algorithm using VQ. Furthermore, to fully use redundant information and to further improve the learning accuracy, an intensified QKMMCC, called QKMMCC_BG, is also proposed on the basis of the bilateral gradient (BG) technique. Simulation results show that, compared with some similar approaches, our algorithms can achieve better computational performance. In addition, our proposed algorithms are also applied to indoor positioning of WiFi-based wireless networks. The experimental results show that our kernel learning algorithms can effectively improve the positioning accuracy. The average positioning errors of our two algorithms in the experiment are 0.86 m and 0.76 m, respectively. The effectiveness of our algorithms is further verified.
暂无评论