Engineering Papers⌕ Search

Engineering topics

Mahjoubfar, Ata

Publications and source records attributed to Mahjoubfar, Ata.

Deep Learning in Label-free Cell Classification

Label-free cell analysis is essential to personalized genomics, cancer diagnostics, and drug development as it avoids adverse effects of staining reagents on cellular viability and cell signaling. However, currently available label-free cell assays mostly rely only on a single feature and lack sufficient differentiation. Also, the sample size analyzed by these assays is limited due to their low throughput. Here, we integrate feature extraction and deep learning with high-throughput quantitative imaging enabled by photonic time stretch, achieving record high accuracy in label-free cell classification. Our system captures quantitative optical phase and intensity images and extracts multiple biophysical features of individual cells. These biophysical measurements form a hyperdimensional feature space in which supervised learning is performed for cell classification. We compare various learning algorithms including artificial neural network, support vector machine, logistic regression, and a novel deep learning pipeline, which adopts global optimization of receiver operating characteristics. As a validation of the enhanced sensitivity and specificity of our system, we show classification of white blood T-cells against colon cancer cells, as well as lipid accumulating algal strains for biofuel production. In conclusion, this system opens up a new path to data-driven phenotypic diagnosis and better understanding of the heterogeneous gene expressions in cells.

60 APPLIED LIFE SCIENCES↗

Ultrafast dark-field surface inspection with hybrid-dispersion laser scanning

High-speed surface inspection plays an important role in industrial manufacturing, safety monitoring, and quality control. It is desirable to go beyond the speed limitation of current technologies for reducing manufacturing costs and opening a new window onto a class of applications that require high-throughput sensing. Here, we report a high-speed dark-field surface inspector for detection of micrometer-sized surface defects that can travel at a record high speed as high as a few kilometers per second. This method is based on a modified time-stretch microscope that illuminates temporally and spatially dispersed laser pulses on the surface of a fast-moving object and detects scattered light from defects on the surface with a sensitive photodetector in a dark-field configuration. The inspector's ability to perform ultrafast dark-field surface inspection enables real-time identification of difficult-to-detect features on weakly reflecting surfaces and hence renders the method much more practical than in the previously demonstrated bright-field configuration. Consequently, our inspector provides nearly 1000 times higher scanning speed than conventional inspectors. To show our method's broad utility, we demonstrate real-time inspection of the surface of various objects (a non-reflective black film, transparent flexible film, and reflective hard disk) for detection of 10 μm or smaller defects on a moving target at 20 m/s within a scan width of 25 mm at a scan rate of 90.9 MHz. Our method holds promise for improving the cost and performance of organic light-emitting diode displays for next-generation smart phones, lithium-ion batteries for green electronics, and high-efficiency solar cells.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

High-speed nanometer-resolved imaging vibrometer and velocimeter

Conventional laser vibrometers are incapable of performing multidimensional vibrometry at high speeds because they build on single-point measurements and rely on beam scanning, significantly limiting their utility and precision. Here we introduce a laser vibrometer that performs high-speed multidimensional imaging-based vibration and velocity measurements with nanometer-scale axial resolution without the need for beam scanning. As a proof-of-concept, we demonstrate real-time microscopic imaging of acoustic vibrations with 1 nm axial resolution, 1200 image pixels, and 30 ps dwell time at 36.7 MHz scan rate.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗