Artificial Intelligence in Label-Free Microscopy: Biological Cell Classification by Time Stretch

ISBN-13: 9783319846545 / Angielski / Miękka / 2018 / 134 str.

Ata Mahjoubfar; Claire Lifan Chen; Bahram Jalali

Opis

This book introduces time-stretch quantitative phase imaging (TS-QPI), a high-throughput label-free imaging flow cytometer developed for big data acquisition and analysis in phenotypic screening. TS-QPI is able to capture quantitative optical phase and intensity images simultaneously, enabling high-content cell analysis, cancer diagnostics, personalized genomics, and drug development. The authors also demonstrate a complete machine learning pipeline that performs optical phase measurement, image processing, feature extraction, and classification, enabling high-throughput quantitative imaging that achieves record high accuracy in label -free cellular phenotypic screening and opens up a new path to data-driven diagnosis.

Spis treści

I Time stretch imaging

1 Introduction

2 Background

2.1 Time Stretch Imaging

2.2 Cell Classification Using Time Stretch Imaging

2.3 Label-free Phenotypic Screening

2.4 Warped Time Stretch for Data Compression

II Inspection and vision

3 Nanometer-resolved imaging vibrometer

3.1 Introduction

3.2 Experimental demonstration

3.3 Theoretical study of the vibrometer performance

3.4 Experimental results

3.5 Conclusion

4 Three-dimensional ultrafast laser scanner

4.1 Introduction

4.2 Principle of hybrid dispersion laser scanner

4.3 Applications of hybrid dispersion laser scanner

III Biomedical applications

5 Label-free High-throughput Phenotypic Screening

5.1 Introduction

5.2 Experimental Setup

5.3 Results and Discussion

5.4 Conclusion

6 Time Stretch Quantitative Phase Imaging

6.1 Background

6.2 Time Stretch Quantitative Phase Imaging

6.2.1 Overview

6.2.2 Imaging system

6.2.3 System performance and resolvable points

6.2.4 Micro

uidic channel design and fabrication

6.2.5 Coherent Detection and Phase Extraction

6.2.6 Cell Transmittance Extraction

6.2.7 Image Reconstruction

6.3 Image Processing Pipeline

6.3.1 Feature Extraction

6.3.2 Multivariate Features Enabled by Sensor Fusion

6.3.3 System Calibration

6.4 Conclusion

IV Big data and artifitial intelligence

7 Big data acquisition and processing in real-time

7.1 Introduction

7.2 Technical description of the acquisition system

7.3 Big data acquisition results

7.4 Conclusion

8 Deep Learning and Classification

8.1 Background

8.2 Machine Learning

8.3 Applications

8.3.1 Blood Screening: Demonstration in Classification of OT-II and SW-480 Cells

8.3.2 Biofuel: Demonstration in Algae Lipid Content Classification

8.4 Further Discussions in Machine Learning

8.4.1 Learning Curves

8.4.2 Principal Component Analysis (PCA)

8.4.3 Cross Validation

8.4.4 Computation Time

8.4.5 Data Cleaning

8.5 Conclusion

V Data compression

9 Optical Data Compression in Time Stretch Imaging

9.1 Background

9.2 Warped Stretch Imaging

9.3 Optical Image Compression

9.4 Experimental Design and Results

9.5 Conclusion

10 Design of Warped Stretch Transform

10.1 Overview

10.2 Kernel Design

10.2.1 Spectral Resolution

10.2.2 Group Delay Profile Design

10.2.3 Simulation Model

10.2.4 Spectrograms

10.3 Discussion

10.4 Conclusion

11 Concluding Remarks and Future Work

References

O autorze

Ata Mahjoubfar is a postdoctoral scholar in the Department of Electrical Engineering and California NanoSystems Institute at University of California Los Angeles. He received his Bachelor's and Master's degrees from University of Tehran, Iran in 2006 and 2008, respectively, and his Ph.D. degree from University of California, Los Angeles (UCLA) in 2014. He was the cofounder of OSA/SPIE student chapter at UCLA and its president in 2012. He is the author of more than 35 peer-reviewed publications, and he holds two international patents. His research interests include artificial intelligence, machine vision and learning, imaging and visualization, ultrafast data acquisition and analytics, biomedical technology, and financial engineering.

Claire Lifan Chen is a senior application engineer at Lumentum Operations LLC. She received her Ph.D. degree in Electrical Engineering and M.Sc. degree in Bioengineering at University of California, Los Angeles in 2015 and 2012, respectively. She received her B.Sc. degree in Optics Science and Engineering from Fudan University, China in 2010. Dr. Chen was a member of California NanoSystems Institute from 2013 to 2015 and the president of OSA/SPIE student chapter at UCLA in 2015. She has authored and co-authored 14 peer-reviewed publications and 2 patents. Her research interests include machine learning, data acquisition and analytics, image processing, and high-throughput imaging with applications in biomedical and information technologies.

Bahram Jalali is the Northrop-Grumman Endowed Chair and Professor of Electrical Engineering at UCLA with joint appointments in Biomedical Engineering, California NanoSystems Institute (CNSI) and Department of Surgery at the UCLA School of Medicine. He is the inventor of the Photonic Time Stretch, a measurement technique that has led to discoveries of new scientific phenomena and to technological inventions. He received his Ph.D. in Applied Physics from Columbia University in 1989 and was with Bell Laboratories in Murray Hill, New Jersey until 2002 before joining UCLA. He is a Fellow of IEEE, the Optical Society of America (OSA), the American Physical Society (APS) and SPIE. He is the recipient of the R.W. Wood Prize from Optical Society of America for the invention and demonstration of the first Silicon Laser, the Aron Kressel Award of the IEEE Photonics Society, the Achievement Medal from IET for his contributions to field of instrumentation for cancer detection, and the Distinguished Engineering Achievement Award from the Engineers Council. He was the founder and CEO of Cognet, a CMOS fiber optics company that was acquired by Intel in 2001. In 2005 he was elected into the Scientific American Top 50, and received the BrideGate 20 Award in 2001 for his entrepreneurial accomplishments and contributions to Southern California economy.

Blurb

This book introduces time-stretch quantitative phase imaging (TS-QPI), a high-throughput label-free imaging flow cytometer developed for big data acquisition and analysis in phenotypic screening. TS-QPI is able to capture quantitative optical phase and intensity images simultaneously, enabling high-content cell analysis, cancer diagnostics, personalized genomics, and drug development. The authors also demonstrate a complete machine learning pipeline that performs optical phase measurement, image processing, feature extraction, and classification, enabling high-throughput quantitative imaging that achieves record high accuracy in label -free cellular phenotypic screening and opens up a new path to data-driven diagnosis.

• Demonstrates how machine learning is used in high-speed microscopy imaging to facilitate medical diagnosis;

• Provides a systematic and comprehensive illustration of time stretch technology;

• Enables multidisciplinary application, including industrial, biomedical, and artificial intelligence.