13:30
Surgery and Intervention
Impact of electrodes orientation on irreversible electroporation
Girindra Wardhana, Nivedha Mamundur Raman, Momen Abayazid, Jurgen J Fütterer
Abstract: Purpose
Over the years, Irreversible Electroporation (IRE) has been developed to specifically destroy cancer tissues as an alternative to surgical resection. In this treatment, placing multiple electrodes in parallel is required to produce a uniform electric field distribution. Parallel electrode placement is a challenging process, and the effect of its accuracy has not been investigated quantitatively. This study investigates the impact of electrode orientation along with various electrode and pulse parameters on the outcomes of IRE.
Methods
The electrode configurations that were considered were parallel, forward, and sideward orientation with a range of 5°-15°. The numerical model was developed to study the effect of electrode orientation on the electric field distribution, which was validated experimentally on potatoes tuber as it has similar properties to biological tissue. In addition, a conductivity test was performed to evaluate the conductivity and the electroporation threshold of the potatoes. Finally, a two-way ANOVA test was selected for the statistical test with the electroporated region as the dependent variable and electrode orientations along with IRE parameters as the independent factors.
Results
The developed numerical model was validated by comparing the electroporated regions between potatoes from experiment and simulation, which achieved a mean dice score of 0.727±0.046. The potato electrical conductivity varies with the increasing electric field given during the electroporation. Using the sigmoid model, we obtained the conductivity value of 0.138-0.473 S/m with an electric field threshold of 250 V/cm. ANOVA test showed that electrode orientations (p<0.001), electrode distance (p<0.001), and active length (p<0.001) had a significant influence on the electroporated region. Further, the difference in the electroporated regions obtained between a parallel orientation and a 5° sideward and forward orientation was not significant (p=0.433).
Conclusions
This study validated the developed numerical models and showed that they predicted the outcome of IRE on potatoes. In addition, a 5° tolerance on the electrode orientation can be defined to obtain a similar response to the parallel orientation.
|
Design and control of a monolithic compliant continuum manipulator for minimally invasive surgery
Theodosia Lourdes Thomas, Jakub Sikorski, G. K. Ananthasuresh, Venkatasubramanian Kalpathy Venkiteswaran, Sarthak Misra
Abstract: Introduction:
Continuum manipulators adapted as steerable catheters can precisely maneuver inside the human body and reach difficult-to-access surgical sites. Compliant design of manipulators coupled with magnetic actuation leads to monolithic compact designs that are easy to sterilize, and offers contactless control. This work presents a new design and control strategy for a magnetically-actuated monolithic compliant continuum manipulator.
Methods:
The continuum manipulator is made from a hollow Titanium tube of outer diameter of 3 mm and wall thickness of 1 mm. It has a novel slotting pattern of flexures with built-in mechanical motion constraints that limit the maximum stress in the flexures. The manipulator can be moved in 3D space by attaching a permanent magnet to its tip and subjecting it to controlled external magnetic fields. A rigid-body model is proposed to analyze the deformation characteristics of the manipulator. Furthermore, a multi-core fiber inscribed with fiber Bragg grating (FBG) sensors is integrated within the manipulator for shape sensing. Closed-loop control of the manipulator to trace different trajectories is demonstrated in an electromagnetic setup. Finally, the steerability of the manipulator is shown in a bifurcating arterial phantom with a miniature camera.
Results and Conclusion:
Static experiments with the manipulator in an electromagnetic setup result in an average error in tip position of 5.27 mm, measured between the rigid-body model and cameras. Closed-loop control of the manipulator tip traces two trajectories: square and straight line. Results show average errors between the desired and estimated trajectory of 4.3 mm, and that between the estimated and camera tracked trajectory of 7.1 mm. An observer-based fusion of rigid-body model and FBG model acts as feedback for the controller. The fusion algorithm offers real-time monitoring and a visualization alternative to X-ray exposure during operation. Furthermore, a miniature camera of 0.91 mm diameter is integrated within the manipulator to demonstrate an endoscopy application. The manipulator is accurately guided inside an undulating tapered channel of a bifurcating arterial phantom. The controlled contactless actuation of the monolithic continuum manipulator with sensor integration shows its clinical feasibility as a steerable catheter or surgical instrument.
|
Optical electrosurgical knife for real-time tumor margin detection
Sara Azizian Amiri, Jenny Dankelman, Benno H. W. Hendriks
Abstract: Introduction: Difficulty in distinguishing the tumor tissue from healthy tissue during lumpectomy surgery can lead to incomplete resection of the tumor and the emergence of a positive margin which may have to be followed by a re-excision surgery or boost radiation therapy. Diffused reflectance spectroscopy (DRS) is a promising technique for real-time tumor margin detection during lumpectomy. Recently it has been shown that the Fat/Water-ratio of the tissues measured using DRS can be used to distinguish breast tumor tissue from healthy tissue [1]. In this study, the function and potential of an electrosurgical knife equipped with DRS (Optical Knife)[2, 3] in detecting tissues during electrosurgery were investigated.
Materials and Methods: For equipping the electrosurgical knife with DRS, a blade-shaped electrode with a metal tube welded on each side was used as optical fiber housing. Moreover, quartz tubes and PTFE tubes were used to protect the optical fibers from high temperatures and debris formation during electrosurgery. A major challenge during the electrosurgery with the Optical Knife is the debris production and its attachment to the quartz tubes that can lead to misdiagnosis of tissue. Different experiments were performed to study the ability of the Optical Knife to detect the type of tissue after using the knife for electrosurgery with different durations and settings (different powers and modes). Furthermore, the potential of the Optical Knife in real-time tissue detection was investigated by performing DRS measurement while cutting through a layered porcine tissue (layers of adipose and muscle tissue) and then a layered phantom (layers with different Fat/Water-ratios). In all experiments, Matlab-based software was used for analyzing DRS outputs and extracting the Fat% and Fat/Water-ratio from each spectrum.
Results: The Fat% and Fat/Water-ratios of tissues measured with the DRS probe were not significantly different from the Fat% and Fat/Water-ratios of the same tissue measured with the used Optical Knife. The tissue type was correctly recognized even when the Optical Knife was used beforehand for electrosurgery for longer durations (14 minutes of cutting) at high power (60W) in pure cutting mode. Furthermore, the measured Fat% and Fat/Water-ratio using the Optical Knife during cutting through the layered porcine tissue and then the layered phantom showed the clear transition between the layers.
Discussions and Conclusions: These results indicate that the optical fibers are well-protected from high temperature and debris formation during electrosurgery with the Optical Knife. It was also possible to cut and detect tissue correctly at the same time using the Optical Knife. These results suggest that the Optical Knife could be a promising tool for real-time tumor margin evaluation during lumpectomy.
References:
1. De Boer, LL, et al. Breast cancer research treatment, 2015.
2. Azizian Amiri, et al. Biomedical Optics Express, 2020.
3. Azizian Amiri, et al. Proceedings of Advanced Biomedical and Clinical Diagnostic and Surgical Guidance Systems XIX. 2021.
|
Ultrasound-guided breast biopsy using tissue displacement tracking and adapted automated cone-based ultrasound scanner which allows 3D US – MRI fusion: A feasibility study
Anton Nikolaev, Leon de Jong, Vincent Groenhuis, Françoise Siepel, Stefano Stramigioli, Hendrik Hansen, Chris de Korte
Abstract: Among available breast biopsy techniques, an ultrasound-guided biopsy is preferable. In case of US occult lesions, MRI-guided biopsy might be performed. We designed a novel Automated Cone-based Breast Ultrasound Scanning and Biopsy System (ACBUS-BS) which is based on the previously developed Automated Cone-based Breast Ultrasound Scanning (ACBUS) system and facilitates biopsy of ultrasound (US) occult lesions in a semi-automated manner.
The biopsy procedure with the ACBUS-BS comprises four steps: target localization, positioning, preparation, and biopsy. At the first step, the target lesion is localized by 3D US-MRI image fusion. At the second step, the system is positioned in the way to support a breast while the needle can be observed on real-time B-mode images during an intervention. The cross-correlation-based tissue displacement tracking algorithm is utilized to estimate the deviation of the localized target from its registered position. The system also provides access to the breast surface to maintain the hygiene of the procedure.
We calculated the errors introduced during each step and their impact on the biopsy outcome as a minimum length of the biopsy sample. For this, we utilize soft polyvinyl alcohol (PAV) custom-made breast phantom (median stiffness is 7.6 kPa) with 8 stained spherical inclusions (10 mm in diameter) of various contrast to noise ratios (14.67+/-0.95 and 3.39+/- 3.15 dB). This phantom was also used for biopsy. Besides, we give an approximation for biopsy of the real breast by utilizing a commercial breast tissue-mimicking phantom and in-vivo measurements.
In the PVA phantom, the mean registration and tracking errors were 1.33 mm and 2.32 mm respectively. For the in-vivo approximation, the mean registration and tracking errors were 2.23 mm and 1.95 mm respectively. The abovementioned errors can lead to biopsy outcomes of 6.81 mm for the phantom, and 5.49 mm for the real breast. After biopsying the phantom, the 6.00+/-0.92 mm of stained material was presented in the biopsy samples.
To our belief, the ACBUS-BS offers a low-cost alternative to MRI-guided biopsy.
|
Quantitative ultrasound and artificial intelligence for staging hepatic steatosis in non-alcoholic fatty liver disease, using conventional ultrasound imaging, validated with histopathology
Gert Weijers, Eric Tjwa, Chris de Korte
Abstract: Background: Nonalcoholic fatty liver disease (NAFLD) is the most common liver disorder in developed countries with a global prevalence of approximately 25%. NAFLD represents a spectrum of disorders and starts with benign steatosis (NAFL, ≥5% hepatic steatosis (HS)), but may lead to nonalcoholic steatohepatitis (NASH), formation of fibrosis, and even cirrhosis. Liver biopsy is still the gold standard for staging steatosis. Biopsies however are invasive and complications such as bleeding and infection may occur, which limits its use as a population screening tool. For this reason, we retrospectively tested a quantitative method using ultrasound (US) B-mode images (Computer Aided UltraSound (CAUS)) , for the non-invasive assessment of HS in a large retrospective human study paralleled by liver biopsy using. Secondly, we explored the potential of artificial intelligence instead to stage HS.
Methods: In this study patients who received a liver biopsy were imaged with transabdominal ultrasound (Siemens Acuson X150; with CH5-2 transducer, ~5 images per patients) generating 913 images from 195 patients. Liver biopsies were qualitatively scored using the Brunt-score for steatosis grading (>5%) and quantitatively (QHIST, fat area %) using in home developed image analysis system working on digitized HE scans (QHIST). For the CAUS analysis, the data was split into a training (50%) and validation (50%) set, and ROC analysis was performed. For training the AI model, a pretrained ImageNet_v2 convolution neural network (CNN) architecture, including connective layers for binary HS classification, was used. As input, the post-processed (i.e. normalized: back-scan converted and beam-profile corrected) CAUS images were used. The data was split into a training and test-set (80-20%) using a central square ROI of 224x224 pixels. Training data again was split (80-20%) for training-validation purpose, while applying data augmentation (vertical flip). Validation was performed on the Area under the curve (AUC) using Adam’s optimizer.
Results: CAUS AUC to QHIST was found to be 0.99 (Sens: 0.96, Spec: 0.94), CAUS AUC to pathologist reading 0.97 (Sens: 0.90, Spec: 0.98). CNN AUC was found to be 0.92 (Sens: 0.88, Spec: 0.75). These findings are in line with previously performed studies in dairy cows. The combination of quantitative histology and quantitative ultrasound showed the best performance.
Conclusion: CAUS is able to classify HS accurately. Usage of AI for HE detection on B-mode ultrasound also seems feasible, however further research is needed to further improve the AI results to comparable level as using quantitative ultrasound.
|
Technology-supported shared decision making in the treatment and management of chronic conditions: a systematic review protocol
Roswita Vaseur, Wendy Oude Nijeweme - d’Hollosy, Monique Tabak
Abstract: Background: In current healthcare, there can be a transition observed towards more preventive and patient-centered care in which technology plays an increasing role. [1-4]. For decision-making specifically this could entail an active role of patients by shared decision making (SDM) and/or active use of technology by i.e. clinical decision support systems (CDSSs) [5, 6]. A systematic review will be conducted to explore how decision-making is structured in both SDM and CDSSs, and to determine how effective they are for adults with a chronic condition. This abstract reports on the protocol that adheres to the PRISMA-P guidelines [7], and will be registered in PROSPERO.
Methods: Studies will be eligible when including: (P) adults with a chronic condition (i.e. COPD, IHD, CHF, diabetes, depression or anxiety), (I) SDM processes or CDSSs that assist in decision-making, (C) control groups receiving usual care, (O) patient-reported, patient-experience or clinical outcomes, and (S) controlled clinical trials. Studies will be identified through PubMed, Scopus, Embase, Web of Science, Cinahl and PsychINFO. Search strings will be conducted in English and limited to studies published from the 1st of January 2011 until the 31st of December 2021. Endnote (Clarivate Analytics, USA) will be used for de-duplication, ASReview (Utrecht University, The Netherlands) for title and abstract screening, and Covidence (Veritas Health Innovation, Australia) for full-text screening, inter-rater reliability, data extraction and risk-of-bias assessment. Finally, RevMan (Nordic Cochrane Centre, Denmark) will be used for data synthesis, if applicable, and data visualization. Two reviewers will independently execute the protocol, whereby for unresolved disagreements a third reviewer will be consulted.
Results: Roles of patients and physicians in SDM and CDSSs are described based on a SDM coding system by Clayman et al., whereas the role of technology in SDM and CDSSs are described based on five primary interaction styles defined by Shneiderman [8, 9]. The effectiveness of SDM and CDSSs are evaluated based on patient and clinical outcomes. Preliminary results are expected in January 2022.
Discussion: This review will allow for interpreting the effects of SDM and CDSSs critically and identify knowledge gaps in technology-supported shared decision making for future research and design.
REFERENCES
1. Thimbleby, H., Technology and the future of healthcare. Journal of public health research, 2013. 2(3).
2. Azodo, I., et al., Opportunities and Challenges Surrounding the Use of Data From Wearable Sensor Devices in Health Care: Qualitative Interview Study. J Med Internet Res, 2020. 22(10): p. e19542.
3. Grady, P.A. and L.L. Gough, Self-management: a comprehensive approach to management of chronic conditions. American journal of public health, 2014. 104(8): p. e25-e31.
4. Tinetti, M.E., et al., Patient Priority-Directed Decision Making and Care for Older Adults with Multiple Chronic Conditions. Clin Geriatr Med, 2016. 32(2): p. 261-75.
5. Abbasgholizadeh Rahimi, S., et al., Are mobile health applications useful for supporting shared decision making in diagnostic and treatment decisions? Glob Health Action, 2017. 10(sup3): p. 1332259.
6. Sutton, R.T., et al., An overview of clinical decision support systems: benefits, risks, and strategies for success. NPJ Digit Med, 2020. 3: p. 17.
7. Shamseer, L., et al., Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ, 2015. 350: p. g7647.
8. Clayman, M.L., et al., Development of a shared decision making coding system for analysis of patient-healthcare provider encounters. Patient Educ Couns, 2012. 88(3): p. 367-72.
9. Shneiderman, B., A taxonomy and rule base for the selection of interaction styles, in Readings in Human–Computer Interaction. 1995, Elsevier. p. 401-410.
|
|