Automatic Segmentation and Probe Guidance for Real-Time Assistance of Ultrasound-Guided Femoral Nerve Blocks

Smistad, Erik; Iversen, Daniel Høyer; Leidig, Linda; Bakeng, Janne Beate Lervik; Johansen, Kaj Fredrik; Lindseth, Frank

doi:10.1016/j.ultrasmedbio.2016.08.036

Cited by 17 publications

(24 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Heterogeneous systems are machines with multiple processors for different purposes such as CPUs, integrated GPUs, dedicated GPUs and vision processing units (VPUs). FAST has been used successfully in several projects such as ultrasound-guided regional anesthesia assistant [4], [5], automatic echocardiography image analysis [6], [7] and the ultrasound robot framework EchoBot [8]. In these projects, FAST is used to create complete software which can handle all necessary steps: data streaming/import/export, processing, visualization, and graphical user interface.…”

Section: Introductionmentioning

confidence: 99%

High Performance Neural Network Inference, Streaming, and Visualization of Medical Images Using FAST

2019

Self Cite

View full text Add to dashboard Cite

Deep convolutional neural networks have quickly become the standard for medical image analysis. Although there are many frameworks focusing on training neural networks, there are few that focus on high performance inference and visualization of medical images. Neural network inference requires an inference engine (IE), and there are currently several IEs available including Intel's OpenVINO, NVIDIA's TensorRT, and Google's TensorFlow which supports multiple backends, including NVIDIA's cuDNN, AMD's ROCm and Intel's MKL-DNN. These IEs only work on specific processors and have completely different application programming interfaces (APIs). In this paper, we presents methods for extending FAST, an open-source high performance framework for medical imaging, to use any IE with a common programming interface. Thereby making it easier for users to deploy and test their neural networks on different processors. This article provides an overview of current IEs and how they can be combined with existing software such as FAST. The methods are demonstrated and evaluated on three performance demanding medical use cases: real-time ultrasound image segmentation, computed tomography (CT) volume segmentation, and patch-wise classification of whole slide microscopy images. Runtime performance was measured on the three use cases with several different IEs and processors. This revealed that the choice of IE and processor can affect performance of medical neural network image analysis considerably. In the most extreme case of processing 171 ultrasound frames, the difference between the fastest and slowest configuration were half a second vs. 24 seconds. For volume processing, using the CPU or the GPU, showed a difference of 2 vs. 53 seconds, and for processing an whole slide microscopy image, the difference was 81 seconds vs. almost 16 minutes. Source code, binary releases and test data can be found online on GitHub at https://github.com/smistad/FAST/.

show abstract

Section: Introductionmentioning

confidence: 99%

High Performance Neural Network Inference, Streaming, and Visualization of Medical Images Using FAST

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although some are time‐consuming, they detect the needle tip with high accuracy . Interestingly, the same algorithms can be trained to detect other linear structures such as nerves and blood vessels, which allows for the ideal needle trajectory of entry to be calculated for median or femoral nerve blocks .…”

Section: Resultsmentioning

confidence: 99%

“…Moreover, advanced tissue recognising technologies can differentiate nerves, ligaments or even intraneural fascicles using electrical impedance, optical coherence tomography (a specialised fiberoptic laser) or high‐frequency US . The next step, combining nerve detection with advanced needle trajectory planning, is already in development for femoral and median nerves .…”

Section: Discussionmentioning

confidence: 99%

Improving needle tip identification during ultrasound-guided procedures in anaesthetic practice

et al. 2017

View full text Add to dashboard Cite

Ultrasound guidance is becoming standard practice for needle-based interventions in anaesthetic practice, such as vascular access and peripheral nerve blocks. However, difficulties in aligning the needle and the transducer can lead to incorrect identification of the needle tip, possibly damaging structures not visible on the ultrasound screen. Additional techniques specifically developed to aid alignment of needle and probe or identification of the needle tip are now available. In this scoping review, advantages and limitations of the following categories of those solutions are presented: needle guides; alterations to needle or needle tip; three- and four-dimensional ultrasound; magnetism, electromagnetic or GPS systems; optical tracking; augmented (virtual) reality; robotic assistance; and automated (computerised) needle detection. Most evidence originates from phantom studies, case reports and series, with few randomised clinical trials. Improved first-pass success and reduced performance time are the most frequently cited benefits, whereas the need for additional and often expensive hardware is the greatest limitation to widespread adoption. Novice ultrasound users seem to benefit most and great potential lies in education. Future research should focus on reporting relevant clinical parameters to learn which technique will benefit patients most in terms of success and safety.

show abstract

“…However, there are also reports in the literature that TABB has no significant effect on improving postoperative pain and analgesia. Griffiths et al found in the study that there was no significant difference in the time between the first press of the controlled analgesia pump, the total dose of opioids, its clinical side effects, and the postoperative visual analogue system scores compared with the blank control group [5]. Belayy et al found that there was no difference in reducing the side effects of opioids and visual analog system scores [6].…”

Section: Introductionmentioning

confidence: 99%

Automatic Positioning and Recognition of Anesthesia Points Based on Ultrasound Image Guidance Technology

Liu

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Local anesthesia has certain harm to the patient, and the wrong position of the anesthesia point may result in poor anesthesia effect. Based on this, this study built an automatic positioning recognition model based on ultrasound image guidance technology. At the same time, this paper combined the actual requirements of anesthesia to establish a wireless positioning system based on ultrasonic infrared, and the system combined ultrasonic propagation characteristics and sensing receiving characteristics. In addition, in the study, this article combined with the state of the anesthesia point to take ultrasound-guided anesthesia. Finally, this paper designed a comparative experiment to study the performance of automatic positioning of anesthesia points based on ultrasound image guidance technology and collected actual data to analyze. Studies have shown that the method proposed in this study has certain clinical effects and can provide theoretical references for subsequent related research.

show abstract

Automatic Segmentation and Probe Guidance for Real-Time Assistance of Ultrasound-Guided Femoral Nerve Blocks

Cited by 17 publications

References 16 publications

High Performance Neural Network Inference, Streaming, and Visualization of Medical Images Using FAST

High Performance Neural Network Inference, Streaming, and Visualization of Medical Images Using FAST

Improving needle tip identification during ultrasound-guided procedures in anaesthetic practice

Automatic Positioning and Recognition of Anesthesia Points Based on Ultrasound Image Guidance Technology

Contact Info

Product

Resources

About