Dataset Information

Can surgical simulation be used to train detection and classification of neural networks?

ABSTRACT: Computer-assisted interventions (CAI) aim to increase the effectiveness, precision and repeatability of procedures to improve surgical outcomes. The presence and motion of surgical tools is a key information input for CAI surgical phase recognition algorithms. Vision-based tool detection and recognition approaches are an attractive solution and can be designed to take advantage of the powerful deep learning paradigm that is rapidly advancing image recognition and classification. The challenge for such algorithms is the availability and quality of labelled data used for training. In this Letter, surgical simulation is used to train tool detection and segmentation based on deep convolutional neural networks and generative adversarial networks. The authors experiment with two network architectures for image segmentation in tool classes commonly encountered during cataract surgery. A commercially-available simulator is used to create a simulated cataract dataset for training models prior to performing transfer learning on real surgical data. To the best of authors' knowledge, this is the first attempt to train deep learning models for surgical instrument detection on simulated data while demonstrating promising results to generalise on real data. Results indicate that simulated data does have some potential for training advanced classification methods for CAI systems.

SUBMITTER: Zisimopoulos O

PROVIDER: S-EPMC5683210 | biostudies-literature | 2017 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Can surgical simulation be used to train detection and classification of neural networks?

Zisimopoulos Odysseas O Flouty Evangello E Stacey Mark M Muscroft Sam S Giataganas Petros P Nehme Jean J Chow Andre A Stoyanov Danail D

Healthcare technology letters 20170914 5

Computer-assisted interventions (CAI) aim to increase the effectiveness, precision and repeatability of procedures to improve surgical outcomes. The presence and motion of surgical tools is a key information input for CAI surgical phase recognition algorithms. Vision-based tool detection and recognition approaches are an attractive solution and can be designed to take advantage of the powerful deep learning paradigm that is rapidly advancing image recognition and classification. The challenge fo ...[more]

PMID: 29184668

Similar Datasets

Project description:High-throughput B-cell sequencing has opened up new avenues for investigating complex mechanisms underlying our adaptive immune response. These technological advances drive data generation and the need to mine and analyze the information contained in these large datasets, in particular the identification of therapeutic antibodies (Abs) or those associated with disease exposure and protection. Here, we describe our efforts to use artificial intelligence (AI)-based image-analyses for prospective classification of Abs based solely on sequence information. We hypothesized that Abs recognizing the same part of an antigen share a limited set of features at the binding interface, and that the binding site regions of these Abs share share common structure and physicochemical property patterns that can serve as a "fingerprint" to recognize uncharacterized Abs. We combined large-scale sequence-based protein-structure predictions to generate ensembles of 3-D Ab models, reduced the Ab binding interface to a 2-D image (fingerprint), used pre-trained convolutional neural networks to extract features, and trained deep neural networks (DNNs) to classify Abs. We evaluated this approach using Ab sequences derived from human HIV and Ebola viral infections to differentiate between two Abs, Abs belonging to specific B-cell family lineages, and Abs with different epitope preferences. In addition, we explored a different type of DNN method to detect one class of Abs from a larger pool of Abs. Testing on Ab sets that had been kept aside during model training, we achieved average prediction accuracies ranging from 71-96% depending on the complexity of the classification task. The high level of accuracies reached during these classification tests suggests that the DNN models were able to learn a series of structural patterns shared by Abs belonging to the same class. The developed methodology provides a means to apply AI-based image recognition techniques to analyze high-throughput B-cell sequencing datasets (repertoires) for Ab classification.

Project description:Conventionally, convolutional neural networks (CNNs) have been used to identify and detect thorax diseases on chest x-ray images. To identify thorax diseases, CNNs typically learn two types of information: disease-specific features and generic anatomical features. CNNs focus on disease-specific features while ignoring the rest of the anatomical features during their operation. There is no evidence that generic anatomical features improve or worsen the performance of convolutional neural networks for thorax disease classification in the current research. As a consequence, the relevance of general anatomical features in boosting the performance of CNNs for thorax disease classification is investigated in this study. We employ a dual-stream CNN model to learn anatomical features before training the model for thorax disease classification. The dual-stream technique is used to compel the model to learn structural information because initial layers of CNNs often learn features of edges and boundaries. As a result, a dual-stream model with minimal layers learns structural and anatomical features as a priority. To make the technique more comprehensive, we first train the model to identify gender and age and then classify thorax diseases using the information acquired. Only when the model learns the anatomical features can it detect gender and age. We also use Non-negative Matrix Factorization (NMF) and Contrast Limited Adaptive Histogram Equalization (CLAHE) to pre-process the training data, which suppresses disease-related information while amplifying general anatomical features, allowing the model to acquire anatomical features considerably faster. Finally, the model that was earlier trained for gender and age detection is retrained for thorax disease classification using original data. The proposed technique increases the performance of convolutional neural networks for thorax disease classification, as per experiments on the Chest X-ray14 dataset. We can also see the significant parts of the image that contribute more for gender, age, and a certain thorax disease by visualizing the features. The proposed study achieves two goals: first, it produces novel gender and age identification results on chest X-ray images that may be used in biometrics, forensics, and anthropology, and second, it highlights the importance of general anatomical features in thorax disease classification. In comparison to state-of-the-art results, the proposed work also produces competitive results.

Project description:Syndromic surveillance detects and monitors individual and population health indicators through sources such as emergency department records. Automated classification of these records can improve outbreak detection speed and diagnosis accuracy. Current syndromic systems rely on hand-coded keyword-based methods to parse written fields and may benefit from the use of modern supervised-learning classifier models. In this paper, we implement two recurrent neural network models based on long short-term memory (LSTM) and gated recurrent unit (GRU) cells and compare them to two traditional bag-of-words classifiers: multinomial naïve Bayes (MNB) and a support vector machine (SVM). The MNB classifier is one of only two machine learning algorithms currently being used for syndromic surveillance. All four models are trained to predict diagnostic code groups as defined by Clinical Classification Software, first to predict from discharge diagnosis, and then from chief complaint fields. The classifiers are trained on 3.6 million de-identified emergency department records from a single United States jurisdiction. We compare performance of these models primarily using the F1 score, and we measure absolute model performance to determine which conditions are the most amenable to surveillance based on chief complaint alone. Using discharge diagnoses, the LSTM classifier performs best, though all models exhibit an F1 score above 96.00. Using chief complaints, the GRU performs best (F1 = 47.38), and MNB with bigrams performs worst (F1 = 39.40). We also note that certain syndrome types are easier to detect than others. For example, chief complaints using the GRU model predicts alcohol-related disorders well (F1 = 78.91) but predicts influenza poorly (F1 = 14.80). In all instances, the RNN models outperformed the bag-of-words classifiers suggesting deep learning models could substantially improve the automatic classification of unstructured text for syndromic surveillance.

Dataset Information

Can surgical simulation be used to train detection and classification of neural networks?

Publications

Can surgical simulation be used to train detection and classification of neural networks?

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets