A contextual detector of surgical tools in laparoscopic videos using deep learning

Namazi, Babak; Sankaranarayanan, Ganesh; Devarajan, Venkat

doi:10.1007/s00464-021-08336-x

Cited by 20 publications

(11 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The flexibility and accessibility of these methods have thus resulted in numerous successful applications in surgical education [15][16][17][18]. Nevertheless, the majority of existing work has focused on minimally invasive techniques; within this area, AI work has been developed for instrument tracking, work-flow analysis, the identification of critical anatomy, and basic surgical skills such as robotic suturing and knot-tying [19][20][21][22][23].…”

mentioning

confidence: 99%

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background Early introduction and distributed learning have been shown to improve student comfort with basic requisite suturing skills. The need for more frequent and directed feedback, however, remains an enduring concern for both remote and in-person training. A previous in-person curriculum for our second-year medical students transitioning to clerkships was adapted to an at-home video-based assessment model due to the social distancing implications of COVID-19. We aimed to develop an Artificial Intelligence (AI) model to perform video-based assessment. Methods Second-year medical students were asked to submit a video of a simple interrupted knot on a penrose drain with instrument tying technique after self-training to proficiency. Proficiency was defined as performing the task under two minutes with no critical errors. All the videos were first manually rated with a pass-fail rating and then subsequently underwent task segmentation. We developed and trained two AI models based on convolutional neural networks to identify errors (instrument holding and knot-tying) and provide automated ratings. Results A total of 229 medical student videos were reviewed (150 pass, 79 fail). Of those who failed, the critical error distribution was 15 knot-tying, 47 instrument-holding, and 17 multiple. A total of 216 videos were used to train the models after excluding the low-quality videos. A k-fold cross-validation (k = 10) was used. The accuracy of the instrument holding model was 89% with an F-1 score of 74%. For the knot-tying model, the accuracy was 91% with an F-1 score of 54%. Conclusions Medical students require assessment and directed feedback to better acquire surgical skill, but this is often time-consuming and inadequately done. AI techniques can instead be employed to perform automated surgical video analysis. Future work will optimize the current model to identify discrete errors in order to supplement video-based rating with specific feedback.

show abstract

mentioning

confidence: 99%

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…‘SurgAI’ dataset consisted of 461 images. Another study focused on surgical tool detection in laparoscopic videos proposing a multi-label classification named LapTool-Net 62 . LapTool-Net exploited the correlations among different tools and tasks using a recurrent convolutional neural (RNN) network.…”

Section: Methodsmentioning

confidence: 99%

Where do we stand in AI for endoscopic image analysis? Deciphering gaps and future directions

Ali

2022

npj Digit. Med.

View full text Add to dashboard Cite

Recent developments in deep learning have enabled data-driven algorithms that can reach human-level performance and beyond. The development and deployment of medical image analysis methods have several challenges, including data heterogeneity due to population diversity and different device manufacturers. In addition, more input from experts is required for a reliable method development process. While the exponential growth in clinical imaging data has enabled deep learning to flourish, data heterogeneity, multi-modality, and rare or inconspicuous disease cases still need to be explored. Endoscopy being highly operator-dependent with grim clinical outcomes in some disease cases, reliable and accurate automated system guidance can improve patient care. Most designed methods must be more generalisable to the unseen target data, patient population variability, and variable disease appearances. The paper reviews recent works on endoscopic image analysis with artificial intelligence (AI) and emphasises the current unmatched needs in this field. Finally, it outlines the future directions for clinically relevant complex AI solutions to improve patient outcomes.

show abstract

“…We don't require domain-specific expertise for our context-aware model, and the basic design has the ability to improve all currently employed approaches. [23] Erfan Maleki , et al(2022) introduced microstructural examinations, hardness and roughness measurements, produced residual stress and relaxation of residual stress, as well as an axial fatigue test, were some of the tests used to arrive at these conclusions. Then, to predict the mechanical characteristics and fatigue behavior of shot-peened material, a DL technique utilized neural networks was employed.…”

Section: Literature Reviewmentioning

confidence: 99%

Enhancement of the Performance and Accuracy of Soil Moisture Data Transmission in IOT

Khattar

Verma²

2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

The need for soil moisture data transmission in IoT is growing thus there is a need to improve the accuracy and performance of such IoT systems. The focus of this research is on soil moisture that is detected by IoT sensors. IoT sensor data is collected and transmitted to the farmer or user. Using deep learning, actuators can take be trained to take remedial action in case of need, such as the sprinkling of water in the agricultural environment. The main focus of research is to measure data and enhance accuracy and performance. Sensors are used to monitor and record every aspect of the environment in their respective settings. The most commonly utilized sensors include proximity sensors, temperature sensors, smoke sensors, etc. Since human population, industry, and energy consumption are all expected to rise in the next decades, environmental monitoring tools and applications must be upgraded to be more accurate and efficient. The detection and measurement of soil moisture is the initial step for smart agriculture. With the use of agricultural control centers, farmers may identify crops in need of treatment and figure out the best quantity of water, fertilizer, and pesticides to administer based on sensor data and imagery input. For farmers, this ensures the soil gets the proper quantity of chemicals for optimal health, while also cutting expenses and minimizing environmental effects, all while reducing waste. The objective of the research work is to improve the accuracy and performance when soil moisture data is transmitted over IoT devices.

show abstract

A contextual detector of surgical tools in laparoscopic videos using deep learning

Cited by 20 publications

References 28 publications

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching

Where do we stand in AI for endoscopic image analysis? Deciphering gaps and future directions

Enhancement of the Performance and Accuracy of Soil Moisture Data Transmission in IOT

Contact Info

Product

Resources

About