We are passionate entrepreneurs, with extensive experience in the Medical device and Healthcare. We seek to improve lives by overcoming one the greatest challenges faced by the healthcare system in the 21st century - that of precision medicine, and specifically in cancer treatment.
In this journey, we collaborate with physicians, researchers and drug development companies in order to implement our ideas where they are most needed.
Our vision is to provide an integrated solution, that will extract a unique, rich data set from a single pathological slide, and will transform this wealth of data into clinically meaningful information.
Our solution will provide a full characterization of the tumor's microenvironment by profiling multiple biomarkers at the cell level while maintaining and learning from the localization context and morphology aspects.
We envision our system as a significant tool in the hands of researchers and pharma companies, accelerating drug development, as well as a companion diagnostic tool, helping pathologists and oncologists in their daily clinical dilemmas.
Our system is based on a novel optical modality, utilizing hyperspectral imaging and machine learning. Using proprietary algorithms, the system enables to map multiple biomarkers on a single slide and to accurately identify each cancer cell. Detailed cell classification is done not only based on morphology but additionally based on the unique expression profile of each cell.
Machine learning algorithms applied to this unpresented, deep characterization of the tumor, will enable better diagnosis, prognosis as well as predicting drug response.
How it works
Spectral data at every pixel
Pentaomix has developed a breakthrough modality for biopsy imaging which enables its users to analyze biopsies with unprecedented accuracy. It is based on a novel optical modality which allows its users to acquire the spectrum at each pixel of the sample in the range of 400-800nm during the scanning of the slide. The spectral information, which cannot be observed with the eye, is made available here for the first time. This unique information allows us to differentiate between several biomarkers on the same slide. Moreover, extensive studies on many different cancer biopsies showed that consistent differences exist between the spectra of normal and cancer cells, allowing high-reliability automatic classification, enabling reliable diagnostics in borderline cases and non-supervised machine learning.
The power of Artificial Intelligence
The system’s integrated AI algorithms allow to put the complex biomarker maps extracted by the system into practical use.
By identifying the unique biomarker profile of each cell, as well as its morphology parameters, the cells can be automatically classified, creating a high-level perspective of the data. The ability to identify individual cancel cells based only on their spectrum, already demonstrated for 8 different cancer types, will save the need for laborious tagging and enable unsupervised machine learning.
The interaction between the immune system cells, normal tissues cells and cancer cells in the tumor's micro-environment can thus be visualized and studied. In retrospective studies, these parameters can then be correlated, using machine learning, with treatments results, creating a predictive power for the system going forward thus paving the way for a companion diagnostic tool that can be readily available in every hospital.
Integrating morphological, genetic & proteomic information
The system allows the unique combination of both spectral information, biomarker expression profiles, cell morphology features as well as overall cell distribution metrics, creating a unique vantage point.
Fusing the perspective of both the expression dimension (through proteomics or genomics, using FISH) and the geometrical dimension (morphology, distributions) extends our ability to comprehend the true complexity of the tumor, beyond the scope of human observation or any existing diagnostic technique. These enhanced capabilities will drive better, faster drug development as well as better diagnosis and drug selection in the clinic.
Prof. Yuval Garini
Co-Founder & CSO
Yuval Garini is a professor of Biophysics and recent past head of the Nano department at Bar Ilan University. Yuval combines cutting-edge research studies in the organization of the genome with novel developments in applications of biomedical science. Among his ground breaking developments, is the delineation of spectral karyotyping which is revolutionizing the field of cytogenetics.
Dr. Boaz Brill
Co-Founder & CEO
Throughout his career, Boaz filled leadership positions in companies that developed advanced monitoring solutions, including VP R&D and CTO of Nova Measuring Instruments which provides monitoring solutions for semiconductor manufacturing and CEO of GluSense which developed an implantable glucose sensor. Boaz holds a Ph.D. in Physics from the Weizmann Institute of Science, and MBA from Bradford University and is a graduate of the IDF Talpiot program.