Skip to main content

Multi-functional Computational Microscopy for Quantitative Cell Tracking

Periodic Reporting for period 1 - MCMQCT (Multi-functional Computational Microscopy for Quantitative Cell Tracking)

Reporting period: 2015-08-01 to 2017-07-31

The invention of the optical microscope in the 16th century has changed the human race’s perception of the world forever. The ability to observe the world at the micron (1/1000-th of a millimeter) level, opened a window to the exploration of fundamental phenomena in biology, physics, materials science and medicine as it became an indispensable tool with direct implications for many aspects of our lives and enabling new inventions and insights. For global health, microscopes are the standard go-to tool in clinics for diagnosis of many types of conditions and diseases, as well as for fundamental research and scientific recreation. A few decades ago, the digital age was introduced to microscopy, with the advent of the digital cameras and the connection of these sensors to computers. This enabled a surge in development of novel measurement techniques, creating a new renaissance for microscopy and enabling automation and real time, longitudinal tracking of events, as they unfold in the microscopic or even nano-scopic (millionth of a millimeter) world.
While these exciting technological advances were the enablers of progress in many fields, the demand for high quality imaging data is ever increasing, influenced by our own perception of modern image quality. However, state of the art optical microscopes are expensive, require a skillful human resource to operate and maintain them, which might create bottlenecks in clinics and research centers, where this resource is shared by numerous laboratories, which ultimately slows down diagnosis and discovery.
In this project, we are developing novel microscopes and computational tools to address the issues of throughput, cost and labor required for biomedical imaging, for a new generation of optical microscopes with the ultimate goal of making biomedical imaging more accessible to clinics, research and recreation.
The work performed so far includes the following:
1) Reducing the number of measurements required to achieve state of the art, clinically relevant, microscopic images of pathology slides, using the lensfree on chip microscope -
Microscopic imaging of pathology slides is one of the standard diagnostic methods for screening various diseases, including cancer. Previously we have demonstrated a high throughput (field of view >20mm^2), compact lens-free holographic based computational imaging microscope, which enables the high-resolution imaging of clinically relevant dense biological samples. Here, we show a comparable quality reconstruction which was obtained from less than 50% of the previously reported nominal number of measurements, using a novel reconstruction algorithm which applies loose sparsity constraint during the iterative reconstruction. The method is demonstrated on pathology slides of breast cancer tissue and Papanicolaou (Pap) smears.

2) Further reducing the number of diffraction patterns required for retrieving the clinically relevant result to a single hologram, using deep learning framework -
Phase recovery is a classical inverse imaging problem with wide range of applications for materials, life sciences and fundamental physics, using diverse radiation sources and detectors in regimes such as X-ray, electron beams and visible light. Numerical inverse methods applied to address the phase retrieval problem has led to numerous scientific discoveries. Here, we demonstrate a deep convolutional neural network for holographic image recovery and phase retrieval, using only a single hologram, acquired in an on-chip holographic microscope configuration. We chose to demonstrate our framework on densely connected biological tissues, such as blood smear, Papanicolaou smear and breast tissue, where classic a-priori object constraints can’t be implemented, and in turn, require multiple acquired holograms that serve as physical constraints for the phase retrieval algorithm. Never the less, following the training of the network, it enables the recovery of such challenging samples from their single acquired hologram, eliminating the self-interference and twin image distortions. Using this framework to recover the specimen’s image also substantially reduces the required computation time, e.g. a field-of-view of ~1mm2 and a resolution of ~0.37µm was inferred ~4 times faster, when compared with the conventional multiple height reconstruction technique. While we first demonstrate sample specific phase recovery networks, that is, networks that were specifically optimized for each sample type, we show that a universal deep network can also be trained with a small increase in computation time. These results provide an exciting new approach for phase recovery and inverse imaging problems.

3) Enhancing microscope images, using an algorithmic framework -
Maximizing imaging throughput, resolution and depth-of-focus is an ever-sought outcome of every microscopic imaging technique and has wide implications on clinical diagnosis and research. Computational imaging techniques have been extensively applied in the last decades, to enhance the information content, such as spatial resolution or extended field-of-view, for imaging experiments, by applying diverse encoding techniques. These techniques usually require addition of extra hardware to the microscope, making assumptions about the object structure or dynamics and applying time intensive algorithms which require sample specific fine tuning of parameters to ensure satisfying results. Here, we demonstrate a significant enhancement of imaging throughput, extended depth-of-field and spatial resolution of an optical microscope using a deep convolutional neural network. The performance enhancement doesn’t require any additional hardware or special design to the microscope. As a first step, the network is being trained with pairs of corresponding images acquired using low and high NA
The results reported in the previous paragraph represent a paradigm shift from current state of the art practices in computational microscopy. Traditionally, to increase metrics such as resolution, depth-of-field, or field-of-view, one should make assumptions about the object being static (or not changing in some domain, under some probing) and use that to encode the information in that static domain to infer the information at the desired domain. Using the tools that we have developed, we show that we can create novel set of tools that learns the statistics between the data and its enhanced version, without making assumptions on its dynamics. In fact, these results will enable a new dynamic assessment of live samples. These results will help to put computational techniques at the forefront of imaging, augmenting it in a seeming-less way in the near future, with equal importance to the optical components.
On chip holographic microscopy of densely connected biological tissue