Periodic Reporting for period 1 - STRATUM (3D Decision Support Tool for Brain Tumor Surgery)
Okres sprawozdawczy: 2023-12-01 do 2025-05-31
Brain and central nervous system cancer was the 12th most common of cancer-related mortality worldwide in 2020 affecting both sexes and all age groups. Notably it was the 2nd leading cause of cancer mortality in young populations (<34 years). Neurosurgery remains one of the primary treatment modalities for brain tumours, where surgeons aim to achieve a gross total resection of the tumour with the aid of neuronavigation, intraoperative ultrasound, and/or fluorescence guided techniques. Despite current advances, neurosurgeons continue to face several challenges during brain tumour surgeries, such as:
i) the difficulty in differentiating tumour margins, since brain tumours (especially gliomas) diffusely infiltrate the surrounding brain tissue;
ii) the brain-shift phenomenon caused by the movement of the brain after performing the craniotomy and starting the resection that affects the accuracy of neuronavigation;
iii) long waiting times for intraoperative pathology consultation that can take up to 45 min;
iv) the lack of tools to ensure complete low-grade tumour resection.
To address these challenges, STRATUM aims to develop a 3D decision support tool for brain surgery guidance and diagnostics integrating augmented reality and multimodal data processing powered by artificial intelligence (AI) algorithms. This tool will function as a point-of-care computing system and will be developed using a co-creation methodology that actively involves end-users and other stakeholders.
STRATUM will pursue the following objectives:
1) To foster advances in personalized medicine based on multimodal data (including the emerging hyperspectral imaging modality) and AI.
2) To increase the intraoperative diagnostic accuracy of brain tumours, improving surgical outcomes and patients’ quality of life.
3) To reduce surgery time with respect to current neurosurgical operation durations.
4) To improve current cost- and energy-efficiency of neurosurgical workflows.
5) To demonstrate the prototype in a two-year clinical study in 3 clinical sites, including an early health technology assessment.
6) To prepare the preliminary business plan and the TRL9 roadmap after the project ending.
An optimized integration and processing of available and new emerging data sources would aid surgeons in timely efficient and correct decision-making in tissue removal. This would maximize the degree of resection while simultaneously minimize the risk of neurological deficits. Moreover, time efficient surgical procedures not only benefit the patients directly by minimizing anaesthesia time and risks of e.g. postoperative infections, but also indirectly by optimizing resources of the health care system.
i) the difficulty in differentiating tumour margins, since brain tumours (especially gliomas) diffusely infiltrate the surrounding brain tissue;
ii) the brain-shift phenomenon caused by the movement of the brain after performing the craniotomy and starting the resection that affects the accuracy of neuronavigation;
iii) long waiting times for intraoperative pathology consultation that can take up to 45 min;
iv) the lack of tools to ensure complete low-grade tumour resection.
To address these challenges, STRATUM aims to develop a 3D decision support tool for brain surgery guidance and diagnostics integrating augmented reality and multimodal data processing powered by artificial intelligence (AI) algorithms. This tool will function as a point-of-care computing system and will be developed using a co-creation methodology that actively involves end-users and other stakeholders.
STRATUM will pursue the following objectives:
1) To foster advances in personalized medicine based on multimodal data (including the emerging hyperspectral imaging modality) and AI.
2) To increase the intraoperative diagnostic accuracy of brain tumours, improving surgical outcomes and patients’ quality of life.
3) To reduce surgery time with respect to current neurosurgical operation durations.
4) To improve current cost- and energy-efficiency of neurosurgical workflows.
5) To demonstrate the prototype in a two-year clinical study in 3 clinical sites, including an early health technology assessment.
6) To prepare the preliminary business plan and the TRL9 roadmap after the project ending.
An optimized integration and processing of available and new emerging data sources would aid surgeons in timely efficient and correct decision-making in tissue removal. This would maximize the degree of resection while simultaneously minimize the risk of neurological deficits. Moreover, time efficient surgical procedures not only benefit the patients directly by minimizing anaesthesia time and risks of e.g. postoperative infections, but also indirectly by optimizing resources of the health care system.
The STRATUM project has a total duration of 60 months (starting on 01 December 2023) and is divided in 7 WPs (Work Packages) with 5 WPs (WP2 to WP6) dedicated to technical and scientific tasks. During the first 18 months (until 31 May 2025), the Consortium has performed several research activities and has accomplish the following main achievements in each technical and scientific WP:
WP2 – Intraoperative Acquisition System and Data Collection
a. The definition of the medical and technological specifications and ethics of artificial intelligence of the STRATUM tool
b. The development of the first STRATUM prototype for HS (Hyperspectral) data acquisition during neurosurgical procedures.
c. The definition of the protocols for clinical data collection and management, including the preparation of the ethics documentation and the development of the environment for data collection.
WP3 - Diagnostic and Integration Algorithm Development
a. The design and development of an initial set of independent processing algorithms based on data obtained in previous projects for starting implementation and acceleration tasks within WP4.
b. The development of a method to simulate the spectral response of the HS acquisition system developed within WP2 by transforming captured HS data with high spectral resolution and evaluating the performance of the classification algorithms using the simulated HS data.
c. The development of anthropomorphic brain phantoms for simulating the anatomical structure and physical properties of a human brain for the design of brain shift compensation algorithms.
WP4 - HPC Processing Platform Development
a. The characterization of the real-time performance requirements and performance bottlenecks for the processing algorithms proposed by WP3.
b. The analysis of the specifications of the critical kernels to be optimized and offloaded to specific HW accelerators.
c. The analysis and descriptions of the PIM approaches and memory system design suitable for the data processing algorithms proposed in WP3.
WP5 - Interactive 3D GUI and Prototype Co-creation and Integration
a. The design of a co-creation protocol to include end-users and other stakeholders in each of the most relevant steps of the development of the STRATUM tool.
b. The conduction of different co-creation sessions to discuss about the goals and preliminary outcomes of STRATUM.
c. The requirements definition and the preliminary design and development of the interactive easy-to-use non-contact 3D graphical user interface of the STRATUM tool for navigating throughout the multimodal data incorporated in the system and the results provided by the processing algorithms.
WP2 – Intraoperative Acquisition System and Data Collection
a. The definition of the medical and technological specifications and ethics of artificial intelligence of the STRATUM tool
b. The development of the first STRATUM prototype for HS (Hyperspectral) data acquisition during neurosurgical procedures.
c. The definition of the protocols for clinical data collection and management, including the preparation of the ethics documentation and the development of the environment for data collection.
WP3 - Diagnostic and Integration Algorithm Development
a. The design and development of an initial set of independent processing algorithms based on data obtained in previous projects for starting implementation and acceleration tasks within WP4.
b. The development of a method to simulate the spectral response of the HS acquisition system developed within WP2 by transforming captured HS data with high spectral resolution and evaluating the performance of the classification algorithms using the simulated HS data.
c. The development of anthropomorphic brain phantoms for simulating the anatomical structure and physical properties of a human brain for the design of brain shift compensation algorithms.
WP4 - HPC Processing Platform Development
a. The characterization of the real-time performance requirements and performance bottlenecks for the processing algorithms proposed by WP3.
b. The analysis of the specifications of the critical kernels to be optimized and offloaded to specific HW accelerators.
c. The analysis and descriptions of the PIM approaches and memory system design suitable for the data processing algorithms proposed in WP3.
WP5 - Interactive 3D GUI and Prototype Co-creation and Integration
a. The design of a co-creation protocol to include end-users and other stakeholders in each of the most relevant steps of the development of the STRATUM tool.
b. The conduction of different co-creation sessions to discuss about the goals and preliminary outcomes of STRATUM.
c. The requirements definition and the preliminary design and development of the interactive easy-to-use non-contact 3D graphical user interface of the STRATUM tool for navigating throughout the multimodal data incorporated in the system and the results provided by the processing algorithms.
We estimate that the STRATUM Point-of-Care system will ensure, at the end of the project (end of year 2028):
a. A quick, detailed, and accurate intraoperative tissue diagnostic;
b. A detailed, structured, and highly personalised neurosurgical guidance;
c. A correct interpretation of all the available information during the surgical procedure via an easy-to-use non-contact 3D GUI.
In this sense, during the first 18 month of the project (until 31 May 2025), STRATUM has achieved the initial results targeting the development of the STRATUM Point-of-Care computing system to be integrated in neurosurgical environments, which will be technically and clinically validated through two clinical studies:
1) The development of the first STRATUM HS acquisition prototype V1.0 for hyperspectral data acquisition in neurosurgery and the preparation and obtention of the Ethics Committee approval of the observational study (STRATUM-OS) protocol.
2) The development of an initial set of algorithms and workflows using previously captured databases.
3) The characterization of the initial set of algorithms and workflows focusing on the performance analysis.
4) The creation of the first version of the non-contact 3D graphical user interface prototype.
5) The elaboration of the co-creation protocol and carry out co-creation sessions for involving end-users and other stakeholders in STRATUM.
6) The dissemination and communication of the goals and results of STRATUM, including 10 scientific papers in open access.
a. A quick, detailed, and accurate intraoperative tissue diagnostic;
b. A detailed, structured, and highly personalised neurosurgical guidance;
c. A correct interpretation of all the available information during the surgical procedure via an easy-to-use non-contact 3D GUI.
In this sense, during the first 18 month of the project (until 31 May 2025), STRATUM has achieved the initial results targeting the development of the STRATUM Point-of-Care computing system to be integrated in neurosurgical environments, which will be technically and clinically validated through two clinical studies:
1) The development of the first STRATUM HS acquisition prototype V1.0 for hyperspectral data acquisition in neurosurgery and the preparation and obtention of the Ethics Committee approval of the observational study (STRATUM-OS) protocol.
2) The development of an initial set of algorithms and workflows using previously captured databases.
3) The characterization of the initial set of algorithms and workflows focusing on the performance analysis.
4) The creation of the first version of the non-contact 3D graphical user interface prototype.
5) The elaboration of the co-creation protocol and carry out co-creation sessions for involving end-users and other stakeholders in STRATUM.
6) The dissemination and communication of the goals and results of STRATUM, including 10 scientific papers in open access.