top of page

An Intraoperative Probe for Brain Tumour Delineation
(Ph.D. Thesis)

Significance: Tumors affecting the brain are known to have a high incidence, mortality, and morbidity in patients. The accuracy and efficacy of tumor resection during surgery strongly influence the effectiveness of chemotherapy and radiation. Hence, a conservative resection can lead to a significant residual tumor, leading to progression and recurrence. In contrast, excessive resection can lead to neurological impairment due to the removal of adjacent normal. Discrimination of tumors from normal is usually based on the color, angiogenesis, hemorrhage, and consistency of the tissues. However, the margins of diffuse tumors are poorly defined as the alterations in color and angiogenesis are subtle even when viewed under a microscope. Hence, various intra-operative techniques are used to map the eloquent areas of the cortex during surgery. These techniques are essential as the pre-operative imaging tools are rendered ineffective due to the brain's plasticity and brain shift. However, various studies have highlighted the limitations of existing intraoperative techniques such as low resolution, difficulties in visualizing low-grade tumors, low sensitivity, long measurement times, high cost of infrastructure, etc. Thus, there is a need to develop an intraoperative tool to augment brain tumor resection to overcome these limitations.

Proposed Solution: We hypothesize that electromechanical characterization (viscoelastic characterization and electrical impedance spectroscopy) can be combined to overcome the limitations of the existing intraoperative tools. An integrated physical properties-based assessment using three different modalities will provide valuable insights for neurosurgical decision-making. We are backed up by a team of expert neurosurgeons and neuropathologists from the National Institute of Mental Health and Neurosciences (NIMHANS) Bangalore.

Figure: Research workflow for the development of the intraoperative probe for brain tumor delineation.

Home: Image

Electrical resistivity study of human brain tissues and tumors

electrical resistivity study.png

Figure: Process flow for the fabrication of the biochip: (a) 4-inch silicon wafer, (b) SiO2 (1µm) grown using thermal oxidation, (c) photoresist spin-coated on oxidized silicon wafer, (d) patterning of photoresist (e) titanium/platinum deposited on the patterned photoresist, (f) microelectrodes realized using lift-off process, (g) biochip dimensions, and (h) scanning electron microscopy (SEM) image of the biochip, (i) platform for studying electrical resistivity of brain tissues (ex vivo) and the plot of mean resistivity of tumor and corresponding normal tissues (SE = Standard Error). 

We have developed a semi-automated system integrated with biochips, an actuation unit, and electronics to measure the electrical resistivity of ex vivo human brain tissues for differentiating normal and tumor. The electrical resistivity of fresh (n = 48), formalin-fixed for one week (n = 48), and long-term (six months) formalin-fixed (n = 27) healthy human brain samples from different anatomical regions and tumor samples (glioma n = 6; fresh, formalin-fixed for one week, and formalin-fixed for six months) were measured using the automated system. The resistivity of glioma (22.4 ± 1.6 Ω-cm) was significantly lesser than the normal region (82 ± 3 Ω-cm) for fresh tissue samples (p = 5e-8). The trend of lower resistivity of glioma compared to normal was preserved after one week and six months of formalin fixation. We also observed the effects of heterogeneity of normal brain tissue and formalin fixation on the electrical properties of tissues. White matter regions were found to have higher resistivity than grey matter regions. The heterogeneity associated with grey matter regions was lower than the white matter regions. Formalin-fixation increased the magnitude of resistivity measured while retaining the trend across different brain regions and tumors. The study shows that the electrical resistivity could potentially be used as an additional biomarker for delineating normal from the tumor. The results led to more extensive studies of the electrical properties of tissues through electrical impedance spectroscopy.

Arjun B. S., Anil Vishnu G. K., Shilpa Rao, Manish Beniwal, and Hardik J. Pandya, “Electrical Phenotyping of Human Brain Tissues: An Automated System for Tumor Delineation.” IEEE Access, 2022. DOI: https://doi.org/10.1109/ACCESS.2022.3149803

Home: Image

Electromechanical characterization of human brain tissues and tumors

TBME.png

Figure: Representative diagram of the study: (a) freshly resected tumor samples from surgery and fresh brain samples from autopsy for the study, (b) the samples examined by a histopathologist and electromechanical characterization using the semi-automated system, and (c) the histopathology images for tumor and normal along with representative results from the electromechanical characterization of tumor and corresponding normal tissues.

The semi-automated system was modified to integrate MEMS-based electromechanical sensors to enable simultaneous electromechanical characterization of tissues. Electrical impedance and viscoelastic characterization of three types of freshly excised gliomas (glioblastoma (GBM), astrocytoma (AST), and oligodendroglioma (OLI)) (N=8 each) and seventeen different normal brain regions (N=6 each) was performed. The electrical impedance of gliomas (462±56Ω) was found to be significantly lower than corresponding normal (1267±515Ω) regions at 100kHz (p=7.46e-11). The difference in the impedance between individual tumor types and corresponding normal regions was also statistically significant (p=1e-8), suggesting accurate tumor delineation. There were distinct differences in the viscoelastic relaxation responses of high-grade and low-grade gliomas. White matter regions demonstrated higher impedance and faster stress relaxation than grey matter regions as a characteristic of their structural composition. We found that simultaneous electromechanical characterization of brain tumors and normal brain tissues can be an effective biomarker for tumor delineation, grading, and understanding of heterogeneity between the brain regions. The observations suggest the potential use of the technology in a clinical setting to achieve gross total resection and improve treatment outcomes by helping surgeons perform real-time risk evaluation during surgery. 

  1. Arjun B. S., Alekya B., Hari R. S., Vikas V., Anita Mahadevan, and Hardik J. Pandya, “Electromechanical Characterization of Human Brain Tissues: A Biomarker for Tumor Delineation.” IEEE Trans Biomed Eng., 2022. DOI: https://doi.org/10.1109/tbme.2022.3171287

  2. Arjun B. S., V S N Sitaramgupta V, Aswin S, Shilpa Rao and Hardik J. Pandya, “A System-based Approach for the Evaluation of Electromechanical Properties of Brain Tumors.” 44th IEEE EMBC International Engineering in Medicine and Biology Conference, Glasgow, Scotland, July 11-15, 2022. DOI: https://doi.org/10.1109/embc48229.2022.9871879.

Video: Demonstration of the semi-automated system for simultaneous electromechanical characterization of human brain tissues and tumors.

Home: Image

3D Printable Application-specific Continuum Robots using Parametric Modelling

Patient Specific.png

Figure: Designing a universal surgical robot that can navigate through complicated pathways also taking into consideration the interpatient variability in the location of tumors is challenging. To solve this challenge, a software platform is created combining the best of parametric modeling and additive manufacturing.

Continuum robots have shown immense potential in medical applications due to their inherent flexibility and adaptability. A novel approach for designing and fabricating patient-specific continuum robots using 3D printing and parametric modeling techniques is proposed. Robots can be tailored to match the patient's anatomy or clinical use case, improving accuracy and performance during medical interventions. The work describes using a parametric design platform, implemented in Rhino software using the Grasshopper plugin, to create complex geometries and manipulate variables in real time. The platform allows for dynamic and responsive design models, offering a flexible and efficient framework for creating patient-specific robot designs. The design parameters are defined based on clinical applications and geometrical limitations, ensuring manufacturability and printability. The proposed approach shows promising results in reducing design and manufacturing time, providing cost-effective production, and enabling customization for personalized healthcare solutions. Moreover, patient-specific robots enhance their potential in various medical applications, including targeted drug delivery, minimally invasive surgeries, and precise anatomical interventions. 

  1. Arjun B. S., Ajay Krishnan A, and Hardik J. Pandya, “MRI-Compatible Patient-specific Continuum Robots using Parametric Modelling.” IEEE-EMBS International Conference on Body Sensor Networks: Sensor and Systems for Digital Health (IEEE BSN 2023), Boston, Massachusetts, USA, October 9 – 11, 2023. 

  2. Arjun B. S., Ajay Krishnan A., and Hardik J. Pandya, “3D Printable Application-Specific Continuum Robots using Parametric Modelling.” 2023 International Conference on Robotics and Automation, London, UK, April 29 - June 2, 2023.

Video: Demonstration of the graphical User Interface (GUI) developed for parametric generation of 3D printable patient-specific continuum robots. 

Stay Tuned for More Updates!

bottom of page