Adapting Deep Learning for Microscopy: Interaction, Application, and Validation

Abstract: Microscopy is an integral technique in biology to study the fundamental components of life visually. Digital microscopy and automation have enabled biologists to conduct faster and larger-scale experiments with a sharp increase in the data generated. Microscopy images contain rich but sparse information, as typically, only small regions in the images are relevant for further study. Image analysis is a crucial tool for biologists in the objective interpretation and extraction of quantitative measurements from microscopy data. Recently, deep learning techniques have shown superior performance in various image analysis tasks. The models learn feature representations from the data by optimizing for a task. However, the techniques require a significant amount of annotated data to perform well. Domain experts are required to annotate microscopy data, making it expensive and time-consuming. The models offer no insight into their prediction, and the learned features are not directly interpretable. This poses challenges to the reliable utilization of the technique in high-trust applications such as drug discovery or disease detection. High data variability in microscopy and poor generalization performance of deep learning models further increase the difficulty in general usage of the technique. The work in this thesis presents frameworks and methods to solve the practical challenges of applying deep learning in microscopy. The application-specific evaluation approaches were presented to validate the approaches, aiming to increase trust in the system. The major contributions of this work are as follows. Papers I and III present human-in-the-loop frameworks for quick adaption of deep learning to new data and for improving models' performance based on human input in visual explanations provided by the model, respectively. Paper II proposes a template-matching approach to improve user interactions in the framework proposed in Paper I. Papers III and IV present architectural modifications in the deep learning models proposed for better visual explanation and image-to-image translation, respectively. Papers IV and V present biologically relevant evaluations of approaches, i.e., analysis of the deep learning models in relation to the biological task.This thesis is aimed towards better utilization and adaptation of the DL methods and techniques to the microscopy data. We show that the annotation burden for the user can be significantly reduced by intuitive annotation frameworks and using contemporary deep-learning paradigms. We further propose architectural modifications in the models to adapt to the requirements and demonstrate the utility of application-specific analysis in microscopy.