Periodic Reporting for period 3 - SOLUS (Smart optical and ultrasound diagnostics of breast cancer)
Okres sprawozdawczy: 2019-11-01 do 2021-10-31
Breast cancer is the most common cancer in women in Europe: About 1/8 women will develop breast cancer and early diagnosis is essential to ensure a high chance of survival. Thus, it is crucial to provide diagnostic tools with high sensitivity for early detection, and high specificity to avoid false-positive results.
Screening mammography is key to reduce mortality rates. At the same time, screening programs result in a significant number of false positive cases, leading to needless additional imaging and invasive procedures like biopsies, with a negative impact on the patient’s quality of life and high burden on healthcare systems: around 50% of positive breast screening outcomes turn out to be false positives and further examination should have been avoided.
The general aim of the project was to support the diagnosis of breast cancer by non-invasively improving the discrimination between borderline malignant and benign lesions.
To reach that goal, SOLUS combined three techniques that provide distinct and complementary pieces of information on breast tissue: B-mode ultrasound (US) scanning to investigate tissue morphology, shear wave elastography to quantify stiffness, and diffuse optical tomography to estimate composition (collagen, water & fat content) and functional blood parameters (blood volume & oxygen saturation). Each of the three techniques has already proven to be useful to some extent, but their combination can make the difference for the non-invasive diagnosis of breast cancer. Furthermore, different from other imaging methods like MRI or PET, US and optical tomography involve no ionizing radiation or contrast agents, and are cheaper and easier to use.
To make the three techniques available for the first time in a single hand-held imaging probe, a very innovative and compact photonics device was developed –the smart optode– a key element in the multimodal SOLUS system, but also a stand-alone device to perform time domain diffuse optical measurements at multiple wavelengths.
Screening mammography is key to reduce mortality rates. At the same time, screening programs result in a significant number of false positive cases, leading to needless additional imaging and invasive procedures like biopsies, with a negative impact on the patient’s quality of life and high burden on healthcare systems: around 50% of positive breast screening outcomes turn out to be false positives and further examination should have been avoided.
The general aim of the project was to support the diagnosis of breast cancer by non-invasively improving the discrimination between borderline malignant and benign lesions.
To reach that goal, SOLUS combined three techniques that provide distinct and complementary pieces of information on breast tissue: B-mode ultrasound (US) scanning to investigate tissue morphology, shear wave elastography to quantify stiffness, and diffuse optical tomography to estimate composition (collagen, water & fat content) and functional blood parameters (blood volume & oxygen saturation). Each of the three techniques has already proven to be useful to some extent, but their combination can make the difference for the non-invasive diagnosis of breast cancer. Furthermore, different from other imaging methods like MRI or PET, US and optical tomography involve no ionizing radiation or contrast agents, and are cheaper and easier to use.
To make the three techniques available for the first time in a single hand-held imaging probe, a very innovative and compact photonics device was developed –the smart optode– a key element in the multimodal SOLUS system, but also a stand-alone device to perform time domain diffuse optical measurements at multiple wavelengths.
We developed a smart optode to perform time domain diffuse optical tomography at 8 wavelengths in the red and near-infrared spectral range (635-1064 nm). Eight optodes were included in a newly developed multimodal probe, which combines the optical tomography capabilities with US. The probe was integrated in a high-end, commercially available US system built by a SOLUS consortium member.
Each component of the optode had to be specifically designed and developed to ensure state-of-the-art performance at a significantly reduced size: 1) the integrated laser driver to generate picosecond pulses (<200 ps) at 8 selected wavelengths, with suitable pulse shape, high average power (>1 mW) and repetition rate (40 MHz); 2) a wide area time-gated single-photon Silicon PhotoMultiplier (SiPM) detector, where the gated acquisition and the capability to control the extension of the active area (up to 8.6 mm2) are needed to reject superficial reflections and manage detected signals varying over orders of magnitude when the source-detector distance is changed in the mm to cm range to collect tomographic data; 3) dedicated acquisition electronics, including a time-to-digital converter and a histogram builder, with 128 channels, channel width of 72 ps, and dead time <100 ns.
All components were integrated into the single optode for a minimum footprint with high collection efficiency.
The handheld multimodal probe was designed to integrate 8 optodes around the US transducer and include water cooling to guarantee reliable performance of the temperature-sensitive optode components. A position sensor is also present. The lasers are Class 1, enabling a safe use for both the operator and the patient without the discomfort of laser safety counter-measures. A full multimodal imaging system was then developed with dedicated software to control operation with the SOLUS probe and with a conventional US probe. The software is capable of multimodal acquisitions, recording patient data, and carrying out a quick analysis of the optical data in real time.
Algorithms were developed to analyze the optical tomographic data collected at multiple wavelengths inside and outside of the lesion, to estimate tissue composition in terms of oxy- and deoxyhemoglobin, water, lipid and collagen content. Advantage is taken of the morphologic information available from B-mode US imaging to guide optical tomography reconstructions, and improve the accuracy of the estimate of lesion properties.
Any development, from the single components, to the smart optode, up to the multimodal probe and the full system were rigorously tested following protocols recognized at European level for performance assessment in diffuse optics (BIT, Medphot, Neuropt). Furthermore, a new protocol was introduced and applied to characterize the tomographic performances of the system, as no specific protocol was available yet. A full kit of bimodal tissue phantoms, suitable for both US and diffuse optical imaging, was developed and exploited. To mimic breast tissue with a lesion, the kit includes homogeneous and heterogeneous phantoms.
Approval for the clinical validation of the SOLUS system was obtained from the Ethics Committee of Ospedale San Raffaele (OSR) and from the Italian Ministry of Health.
The planned clinical validation is presently ongoing at OSR. It includes: 1) mock sessions to train the physicians on the use of the system and get feedback from them on its usability; 2) an initial test of the diagnostic potential (with data collected from 20 patients with malignant and 20 with benign breast lesions).
Due to the COVID pandemic, the start of the clinical validation was delayed. The SOLUS consortium has agreed to voluntarily continue the validation beyond the official end of the project to fully estimate the diagnostic potential.
Each component of the optode had to be specifically designed and developed to ensure state-of-the-art performance at a significantly reduced size: 1) the integrated laser driver to generate picosecond pulses (<200 ps) at 8 selected wavelengths, with suitable pulse shape, high average power (>1 mW) and repetition rate (40 MHz); 2) a wide area time-gated single-photon Silicon PhotoMultiplier (SiPM) detector, where the gated acquisition and the capability to control the extension of the active area (up to 8.6 mm2) are needed to reject superficial reflections and manage detected signals varying over orders of magnitude when the source-detector distance is changed in the mm to cm range to collect tomographic data; 3) dedicated acquisition electronics, including a time-to-digital converter and a histogram builder, with 128 channels, channel width of 72 ps, and dead time <100 ns.
All components were integrated into the single optode for a minimum footprint with high collection efficiency.
The handheld multimodal probe was designed to integrate 8 optodes around the US transducer and include water cooling to guarantee reliable performance of the temperature-sensitive optode components. A position sensor is also present. The lasers are Class 1, enabling a safe use for both the operator and the patient without the discomfort of laser safety counter-measures. A full multimodal imaging system was then developed with dedicated software to control operation with the SOLUS probe and with a conventional US probe. The software is capable of multimodal acquisitions, recording patient data, and carrying out a quick analysis of the optical data in real time.
Algorithms were developed to analyze the optical tomographic data collected at multiple wavelengths inside and outside of the lesion, to estimate tissue composition in terms of oxy- and deoxyhemoglobin, water, lipid and collagen content. Advantage is taken of the morphologic information available from B-mode US imaging to guide optical tomography reconstructions, and improve the accuracy of the estimate of lesion properties.
Any development, from the single components, to the smart optode, up to the multimodal probe and the full system were rigorously tested following protocols recognized at European level for performance assessment in diffuse optics (BIT, Medphot, Neuropt). Furthermore, a new protocol was introduced and applied to characterize the tomographic performances of the system, as no specific protocol was available yet. A full kit of bimodal tissue phantoms, suitable for both US and diffuse optical imaging, was developed and exploited. To mimic breast tissue with a lesion, the kit includes homogeneous and heterogeneous phantoms.
Approval for the clinical validation of the SOLUS system was obtained from the Ethics Committee of Ospedale San Raffaele (OSR) and from the Italian Ministry of Health.
The planned clinical validation is presently ongoing at OSR. It includes: 1) mock sessions to train the physicians on the use of the system and get feedback from them on its usability; 2) an initial test of the diagnostic potential (with data collected from 20 patients with malignant and 20 with benign breast lesions).
Due to the COVID pandemic, the start of the clinical validation was delayed. The SOLUS consortium has agreed to voluntarily continue the validation beyond the official end of the project to fully estimate the diagnostic potential.
The SOLUS system can achieve substantially improved breast cancer diagnosis, sparing unnecessary additional examinations, including biopsies, that are now performed after a false positive mammography.
The system also allows more effective treatment and therapy management, through better capabilities of characterising tumors. For example, efficient chemotherapy response monitoring and prediction is currently not available, while it would enable personalised decision-making, therapy planning and optimisation for each patient. That would also contribute to a significant decrease of total costs of breast cancer management.
The multimodal SOLUS system can find application in other medical imaging domains as well, such as screening for musculoskeletal diseases, thyroid cancer etc.
The smart optode itself is a ground-breaking, stand-alone device for time domain multi-wavelength diffuse optical measures with potential use in several fields, from wearable devices (e.g. to monitor athletic training) to the non-destructive assessment of wood quality for industrial uses.
The single photonics components (both laser driver & detector) also represent a technology breakthrough beyond the current state-of-the-art.
With the SOLUS protocol and its tissue phantoms, the project has laid the initial foundation of future standards for diffuse optics.
Furthermore, the project brings together key industrial players in the field and contribute to a secured and reinforced industrial leadership for Europe in the biophotonics market. Specifically, two patent applications were filed on the detector and multimodal probe.
The system also allows more effective treatment and therapy management, through better capabilities of characterising tumors. For example, efficient chemotherapy response monitoring and prediction is currently not available, while it would enable personalised decision-making, therapy planning and optimisation for each patient. That would also contribute to a significant decrease of total costs of breast cancer management.
The multimodal SOLUS system can find application in other medical imaging domains as well, such as screening for musculoskeletal diseases, thyroid cancer etc.
The smart optode itself is a ground-breaking, stand-alone device for time domain multi-wavelength diffuse optical measures with potential use in several fields, from wearable devices (e.g. to monitor athletic training) to the non-destructive assessment of wood quality for industrial uses.
The single photonics components (both laser driver & detector) also represent a technology breakthrough beyond the current state-of-the-art.
With the SOLUS protocol and its tissue phantoms, the project has laid the initial foundation of future standards for diffuse optics.
Furthermore, the project brings together key industrial players in the field and contribute to a secured and reinforced industrial leadership for Europe in the biophotonics market. Specifically, two patent applications were filed on the detector and multimodal probe.