Sight OLO

The Sight OLO is an automated hematology analyzer that utilizes a computer vision platform with fluorescence microscopy and image processing algorithms to provide quantitative analysis of 19 complete blood count parameters from small volume whole blood samples. Its design aims to deliver rapid and accessible CBC results with a 5-part differential by automating the identification and characterization of blood cells based on their visual morphology.

The Sight OLO is a quantitative, multi-parameter automated hematology analyzer intended for in vitro diagnostic use within clinical laboratories. It is designed to screen either capillary or venous whole blood samples to enumerate 19 complete blood count (CBC) parameters, which include a 5-part white blood cell (WBC) differential, and is capable of identifying and classifying the formed elements of blood in patients aged three months and older. The device employs a computer vision-based platform, utilizing advanced algorithms to process images of stained blood samples acquired via an integrated fluorescence microscope. This technology automates the identification and quantification of various blood components (such as red blood cells, WBCs, and platelets) and their distinct characteristics—including cell volume, shape, intensity, and morphology—facilitating rapid and straightforward operation through a compact scanning and analyzing unit paired with a test kit containing disposable cartridges and sample preparation tools. The system is also equipped to signal specific WBC abnormalities by flagging samples. Comprehensive performance evaluations have substantiated the Sight OLO's analytical capabilities. Method comparison studies, encompassing 679 diverse patient samples, demonstrated robust correlation and acceptable bias when compared to an established predicate device across all 19 CBC parameters. Rigorous repeatability and reproducibility studies, conducted at multiple sites with varied operators and instruments, affirmed the device's high precision, with all measured parameters consistently meeting predefined acceptance criteria for coefficient of variation. Detection limit assessments established a Limit of Quantitation (LoQ) of 0.18 x 10³/µL for WBC, 13.40 x 10³/µL for platelets, and 3.9 g/dL for hemoglobin, alongside well-defined linearity ranges for these key analytes. Analytical specificity investigations revealed minimal interference from common endogenous and exogenous substances at clinically significant concentrations and confirmed robustness against challenging sample conditions like RBC fragments or extreme cell counts. Sample stability was validated for up to 8 hours at room temperature for most parameters (RDW being 4 hours), and the accompanying test kit has an initial shelf-life of 6 months. Furthermore, flagging studies achieved a sensitivity of 93.0% and a specificity of 80.6% in detecting sample abnormalities relative to manual microscopy. Reference intervals for adult populations were successfully established, pediatric intervals were verified, and matrix comparison studies confirmed the interchangeability of results from capillary and venous samples, as well as different capillary collection techniques.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the computer vision algorithm differentiate between morphologically similar but distinct cell types (e.g., reactive lymphocytes vs. blasts) with high accuracy, especially in complex pathological samples?; What is the typical turnaround time from sample loading to result availability, and how does this compare to traditional laboratory analyzers in high-throughput settings?; Given the 80.6% specificity in flagging studies, what are the most common causes of false positive flags, and what strategies are in place or could be developed to improve this specificity without compromising sensitivity?; The device is factory calibrated; what is the recommended frequency or criteria for verification of this calibration in the field to ensure continued accuracy over the device's lifespan?; How does the performance of the Sight OLO, particularly its computer vision analysis, vary with samples exhibiting extreme hematological abnormalities (e.g., very high leukocytosis, severe anemia, or rare cell morphologies) beyond those explicitly tested in the interference studies?