 |
 |
24 Aug 2021
Seoul startup has developed laser-induced plasma spectroscopy scanner for biomedical use.
Speclipse, a startup company in Korea with a laser-based approach to rapid screening for the most dangerous form of skin cancer, has closed a round of venture funding worth $7.7 million.
The Seoul-based firm says that the laser-induced plasma spectroscopy (LIPS) technique used in its “Spectra-Scope” product is able to give an accurate diagnosis in real time - meaning that lots of unnecessary and invasive tissue biopsies could be avoided.
Skin and blood
Currently setting up a sales network across 20 countries, Speclipse adds that it now wants to extend the use of the non-invasive tool to blood analysis.
“The company's laser spectroscopic technology overcomes the limitations of existing blood analysis technologies in that it can simultaneously detect small variances in various blood biomarkers,” claims the company.
Speclipse adds that similar deep-learning algorithmic technologies to that developed for skin tissue analysis could now be applied to large amounts of blood spectroscopic data, potentially enabling the early diagnosis of cancers in the gastro-intestinal tract, and even Alzheimer's disease.
According to the “Born2Global Centre”, a startup incubator in Korea of which Speclipse is a member, the latest funding brings total investment in the firm so far to $13 million.
Major shareholders in the company, which was set up by CEO and Stanford graduate Sung Hyun Pyun, include Korea-based backers Signite Partners, BNH Investment, Shinhan Capital, and MiCo.
Neural network
More widely known as laser-induced breakdown spectroscopy (LIBS), the LIPS technique has previously been used in several other applications - from profiling molten steel, to analyzing rocks on Mars and at the bottom of the Pacific Ocean in situ.
It works by firing pulses from a nanosecond laser at the target material, thus creating a breakdown plasma with very little associated heat. The generated plasma contains elements of the target material at elevated energy levels, which emit light at characteristic wavelengths as they decay.
Speclipse takes advantage of that phenomenon by using the emission from a Q-switched Nd:YAG laser, in combination with a neural network previously “trained” through biopsy results, to extract biomolecular information from a skin lesion without damaging or removing the tissue.
In a recent study involving 364 patients in Australia it claimed a sensitivity of close to 98 per cent and a sensitivity of 86 per cent for the Spectra-Scope, meaning that although generally good at spotting cancerous lesions it did also produce a number of false-positive results.
“Spectra-Scope’s deep-learning-based proprietary diagnostic algorithm differentiates the conditions of BCC [basal cell carcinoma], SCC [squamous cell carcinoma], and malignant melanoma with high accuracy,” states the firm.
“When doctors find a suspicious skin lesion, they tend to examine it with a dermascope and then decide whether to perform a biopsy.
“However, there are cases where the doctor - because of his or her skill level - makes the wrong decision on whether the lesion needs to be biopsied. Speclipse's technology can be used to avoid these incorrect diagnoses.”
 |  |  |  |  |  |  |
| © 2024 SPIE Europe |
|
