Kelvin C. M. Lee

University of Hong Kong

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Kevin K. Tsia

University of Hong Kong

Dickson M. D. Siu

University of Hong Kong

Kenneth K. Y. Wong

University of Hong Kong

Queenie T. K. Lai

University of Hong Kong

Michelle C. K. Lo

Nano and Advanced Materials Institute

Hayden Kwok‐Hay So

University of Hong Kong

Bob M. F. Chung

Nano and Advanced Materials Institute

Maolin Wang

Capital University of Economics and Business

Anson H. L. Tang

University of Hong Kong

Gwinky G. K. Yip

University of Hong Kong

Cooperative Institutions

University of Hong Kong

Chinese University of Hong Kong

Queen Mary Hospital

Nano and Advanced Materials Institute

HKU-Pasteur Research Pole

Hong Kong Science and Technology Parks Corporation

University of Hong Kong - Shenzhen Hospital

Hong Kong Sanatorium and Hospital

Hong Kong University of Science and Technology

Bellingham Technical College

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Information‐Distilled Generative Label‐Free Morphological Profiling Encodes Cellular Heterogeneity

Advanced Science (2024)

Michelle C. K. Lo Dickson M. D. Siu Kelvin C. M. Lee Justin S. J. Wong Maximus C.F. Yeung

Abstract Image‐based cytometry faces challenges due to technical variations arising from different experimental batches and conditions, such as differences in instrument configurations or image acquisition protocols, impeding genuine biological interpretation of cell morphology. Existing solutions, often necessitating extensive pre‐existing data knowledge or control samples across batches, have proved limited, especially with complex cell image data. To overcome this, “Cyto‐Morphology Adversarial Distillation” (CytoMAD), a self‐supervised multi‐task learning strategy that distills biologically relevant cellular morphological information from batch variations, is introduced to enable integrated analysis across multiple data batches without complex data assumptions or extensive manual annotation. Unique to CytoMAD is its “morphology distillation”, symbiotically paired with deep‐learning image‐contrast translation—offering additional interpretable insights into label‐free cell morphology. The versatile efficacy of CytoMAD is demonstrated in augmenting the power of biophysical imaging cytometry. It allows integrated label‐free classification of human lung cancer cell types and accurately recapitulates their progressive drug responses, even when trained without the drug concentration information. CytoMAD also allows joint analysis of tumor biophysical cellular heterogeneity, linked to epithelial‐mesenchymal plasticity, that standard fluorescence markers overlook. CytoMAD can substantiate the wide adoption of biophysical cytometry for cost‐effective diagnosis and screening.

Profiling (computer programming)

Cytometry

10.1002/advs.202307591

Cite

Citations (2)

Multi‐ATOM: Ultrahigh‐throughput single‐cell quantitative phase imaging with subcellular resolution

Journal of Biophotonics (2019)

Kelvin C. M. Lee Andy K. S. Lau Anson H. L. Tang Maolin Wang Aaron T. Mok

A growing body of evidence has substantiated the significance of quantitative phase imaging (QPI) in enabling cost-effective and label-free cellular assays, which provides useful insights into understanding the biophysical properties of cells and their roles in cellular functions. However, available QPI modalities are limited by the loss of imaging resolution at high throughput and thus run short of sufficient statistical power at the single-cell precision to define cell identities in a large and heterogeneous population of cells-hindering their utility in mainstream biomedicine and biology. Here we present a new QPI modality, coined multiplexed asymmetric-detection time-stretch optical microscopy (multi-ATOM) that captures and processes quantitative label-free single-cell images at ultrahigh throughput without compromising subcellular resolution. We show that multi-ATOM, based upon ultrafast phase-gradient encoding, outperforms state-of-the-art QPI in permitting robust phase retrieval at a QPI throughput of >10 000 cell/sec, bypassing the need for interferometry which inevitably compromises QPI quality under ultrafast operation. We employ multi-ATOM for large-scale, label-free, multivariate, cell-type classification (e.g. breast cancer subtypes, and leukemic cells vs peripheral blood mononuclear cells) at high accuracy (>94%). Our results suggest that multi-ATOM could empower new strategies in large-scale biophysical single-cell analysis with applications in biology and enriching disease diagnostics.

Phase imaging

Cytometry

10.1002/jbio.201800479

Cite

Citations (41)

Toward Deep Biophysical Cytometry: Prospects and Challenges

Trends in biotechnology (2021)

Kelvin C. M. Lee Jochen Guck Keisuke Goda Kevin K. Tsia

Mass cytometry

Cytometry

10.1016/j.tibtech.2021.03.006

Cite

Citations (43)

Multimodal FACED imaging flow cytometry for correlative single-cell morphological analysis

Biophotonics Congress 2021 (2021)

Gwinky G. K. Yip Michelle C. K. Lo Kelvin C. M. Lee Queenie T. K. Lai Kenneth K. Y. Wong

We report a high-throughput multimodal imaging flow cytometer for biophysical and morphological analysis of cell cycle progression at single-cell precision with molecular specificity (at >10,000 cells/sec).

Correlative

Single-Cell Analysis

Cytometry

10.1364/ntm.2021.ntu1c.5

Cite

Citations (0)

Multimodal FACED imaging for large-scale single-cell morphological profiling

APL Photonics (2021)

Gwinky G. K. Yip Michelle C. K. Lo Wenwei Yan Kelvin C. M. Lee Queenie T. K. Lai

Free-space angular-chirp-enhanced delay (FACED) is an ultrafast laser-scanning technique that allows for high imaging speed at the scale orders of magnitude greater than the current technologies. However, this speed advantage has only been restricted to bright-field and fluorescence imaging—limiting the variety of image contents and hindering its applicability in image-based bioassay, which increasingly demands rich phenotypic readout at a large scale. Here, we present a new high-speed quantitative phase imaging (QPI) based on time-interleaved phase-gradient FACED image detection. We further integrate this system with a microfluidic flow cytometer platform that enables synchronized and co-registered single-cell QPI and fluorescence imaging at an imaging throughput of 77 000 cells/s with sub-cellular resolution. Combined with deep learning, this platform empowers comprehensive image-based profiling of single-cell biophysical phenotypes that can offer not only sufficient label-free power for cell-type classification but also cell-cycle phase tracking with high accuracy comparable to the gold-standard fluorescence method. This platform further enables correlative, compartment-specific single-cell analysis of the spatially resolved biophysical profiles at the throughput inaccessible with existing QPI methods. The high imaging throughput and content given by this multimodal FACED imaging system could open new opportunities in image-based single-cell analysis, especially systematic analysis that correlates the biophysical and biochemical information of cells, and provide new mechanistic insights into biophysical heterogeneities in many biological processes.

Phase imaging

Profiling (computer programming)

Fluorescence-lifetime imaging microscopy

10.1063/5.0054714

Cite

Citations (15)

Large-scale single-cell fractal analysis based on quantitative phase imaging

Ziqi Zhang Queenie T. K. Lai Kelvin C. M. Lee Kenneth K. Wong Kevin K. Tsia

We report a large-scale light-scattering single-cell characterization enabled by a high-throughput quantitative phase imaging platform (multi-ATOM) (10,000 cells/sec). By virtue of its subcellular resolution, multi-ATOM accesses the light-scattering information from individual cells via Fourier Transform light scattering (FTLS) analysis. Specifically, we applied FTLS analysis on multi-ATOM images to explore the statistical characteristics of single-cell fractal dimension (FD). We demonstrated that FD can be harnessed as an effective label-free phenotype that is indicative of cell types and states. FD can identify the heterogeneity among leukemia/lung cancer cell types and trace the different phases in cell cycle progression.

Phase imaging

10.1117/12.2583204

Cite

Citations (0)

Optofluidic imaging meets deep learning: from merging to emerging

Lab on a Chip (2023)

Dickson M. D. Siu Kelvin C. M. Lee Bob M. F. Chung Justin S. J. Wong Guoan Zheng

We discuss the recent trends in integrating deep-learning (DL) and optofluidic imaging. A holistic understanding of them could incentivize DL-powered optofluidic imaging for advancing a wide range of novel applications in science and biomedicine.

Biomedicine

10.1039/d2lc00813k

Cite

Citations (14)

Deep-learning-assisted biophysical imaging cytometry at massive throughput delineates cell population heterogeneity

Lab on a Chip (2020)

Dickson M. D. Siu Kelvin C. M. Lee Michelle C. K. Lo Shobana V. Stassen Maolin Wang

The association of the intrinsic optical and biophysical properties of cells to homeostasis and pathogenesis has long been acknowledged. Defining these label-free cellular features obviates the need for costly and time-consuming labelling protocols that perturb the living cells. However, wide-ranging applicability of such label-free cell-based assays requires sufficient throughput, statistical power and sensitivity that are unattainable with current technologies. To close this gap, we present a large-scale, integrative imaging flow cytometry platform and strategy that allows hierarchical analysis of intrinsic morphological descriptors of single-cell optical and mass density within a population of millions of cells. The optofluidic cytometry system also enables the synchronous single-cell acquisition of and correlation with fluorescently labeled biochemical markers. Combined with deep neural network and transfer learning, this massive single-cell profiling strategy demonstrates the label-free power to delineate the biophysical signatures of the cancer subtypes, to detect rare populations of cells in the heterogeneous samples (10-5), and to assess the efficacy of targeted therapeutics. This technique could spearhead the development of optofluidic imaging cell-based assays that stratify the underlying physiological and pathological processes based on the information-rich biophysical cellular phenotypes.

Cytometry

Live cell imaging

10.1039/d0lc00542h

Cite

Citations (47)

Label-Free Phytoplankton Analysis by High-Throughput Quantitative Phase Imaging Cytometry and Machine Learning

Queenie T. K. Lai Kelvin C. M. Lee Kenneth K. Wong Kevin K. Tsia

Phytoplankton, highly diversified in species, plays a critical role in the atmospheric carbon cycle and marine ecosystem [1]. Capable of doing large-scale analysis of phytoplankton is thus of significance in environmental monitoring and biofuel production. However, current techniques lack the throughput to efficiently screen the highly heterogeneous population of phytoplankton at the single-cell precision for accurate taxonomic classification and/or overwhelmingly rely upon biochemical assays, which are typically non-invasive and destructive, to characterise the physiological states and functions. Here we report a high-throughput, label-free imaging flow cytometer (>10,000 cells/sec) based on a new type of quantitative phase imaging technique, called multiplexed asymmetric-detection time-stretch optical microscopy (multi-ATOM) [2] combined with a supervised learning strategy. We demonstrate that multi-ATOM provides sufficient label-free statistical power for multi-class classification of phytoplankton (13 classes) (Fig. 1a-b). We also show that the label-free single-cell biophysical phenotypes (>100) derived from multi-ATOM are sensitive to changes in cultivation nutrient and could be the potential label-free biomarkers for lipid-content screening in biofuel production (Fig. 1c-d).

Cytometry

10.1109/cleoe-eqec.2019.8873409

Cite

Citations (0)

Low-Latency In Situ Image Analytics With FPGA-Based Quantized Convolutional Neural Network

IEEE Transactions on Neural Networks and Learning Systems (2021)

Maolin Wang Kelvin C. M. Lee Bob M. F. Chung Sharatchandra Varma Bogaraju Ho-Cheung Ng

Real-time in situ image analytics impose stringent latency requirements on intelligent neural network inference operations. While conventional software-based implementations on the graphic processing unit (GPU)-accelerated platforms are flexible and have achieved very high inference throughput, they are not suitable for latency-sensitive applications where real-time feedback is needed. Here, we demonstrate that high-performance reconfigurable computing platforms based on field-programmable gate array (FPGA) processing can successfully bridge the gap between low-level hardware processing and high-level intelligent image analytics algorithm deployment within a unified system. The proposed design performs inference operations on a stream of individual images as they are produced and has a deeply pipelined hardware design that allows all layers of a quantized convolutional neural network (QCNN) to compute concurrently with partial image inputs. Using the case of label-free classification of human peripheral blood mononuclear cell (PBMC) subtypes as a proof-of-concept illustration, our system achieves an ultralow classification latency of 34.2 [Formula: see text] with over 95% end-to-end accuracy by using a QCNN, while the cells are imaged at throughput exceeding 29 200 cells/s. Our QCNN design is modular and is readily adaptable to other QCNNs with different latency and resource requirements.

10.1109/tnnls.2020.3046452

Cite

Citations (16)