<p dir="ltr">Generative Image-to-Image Translation (I2IT) involves transforming an input image from one domain to another. Typically, this transformation retains the content in the input image while adjusting the domain-dependent style elements. Generative I2IT finds utility in a wide range of applications, yet its effectiveness hinges on adaptations to the unique characteristics of the data at hand. This dissertation pushes the boundaries of I2IT by applying it to stain-related problems in computational pathology. Particularly, the main contributions span two major applications of stain translation: H&E-to-H&E and H&E-to-IHC, each with its unique requirements and challenges. More specifically, the first contribution addresses the generalization challenge posed by the high variability in H&E stain appearances to any task-specific machine learning models. To this end, the Generative Stain Augmentation Network (G-SAN) is introduced to augment the training images in any downstream task with random and diverse H&E stain appearances. Experimental results demonstrate G-SAN’s ability to enhance model generalization across stain variations in downstream tasks. The second key contribution in this dissertation focuses on H&E-to-IHC stain translation. The major challenge in learning accurate H&E-to-IHC stain translation is the frequent and sometimes severe inconsistencies in the groundtruth H&E-IHC image pairs. To make training more robust to these inconsistencies, a novel contrastive learning based loss, named the Adaptive Supervised PatchNCE (ASP) loss is presented. Experimental results suggest that the proposed ASP-based framework outperforms the state-of-the-art in H&E-to-IHC stain translation by significant margins. Additionally, a new dataset for H&E-to-IHC translation – the Multi-IHC Stain Translation (MIST) dataset, is released to the public, featuring paired images from H&E to four different IHC stains. For future directions of generative I2IT in stain translation problems, a proof-of-concept study of applying the latest diffusion model based I2IT methods to the problem of virtual H&E staining is presented.</p>

<p dir="ltr">X-ray computed tomography (CT) is one of the most widely used imaging modalities for medical diagnosis. Improving the quality of clinical CT images while keeping the X-ray dosage of patients low has been an active area of research. Recently, there have been two major technological advances in the commercial CT systems. The first is the use of Deep Neural Networks (DNN) to denoise and sharpen CT images, and the second is use of photon counting detectors (PCD) which provide higher spectral and spatial resolution compared to the conventional energy-integrating detectors. While both techniques have potential to improve the quality of CT images significantly, there are still challenges to improve the quality further.</p><p dir="ltr"><br></p><p dir="ltr">A denoising or sharpening algorithm for CT images must retain a favorable texture which is critically important for radiologists. However, commonly used methodologies in DNN training produce over-smooth images lacking texture. The lack of texture is a systematic error leading to a biased estimator.</p><p><br></p><p dir="ltr">In the first portion of this thesis, we propose three algorithms to reduce the bias, thereby to retain the favorable texture. The first method proposes a novel approach to designing a loss function that penalizes bias in the image more while training a DNN, producing more texture and detail in results. Our experiments verify that the proposed loss function outperforms the commonly used mean squared error loss function. The second algorithm proposes a novel approach to designing training pairs for a DNN-based sharpener. While conventional sharpeners employ noise-free ground truth producing over-smooth images, the proposed Noise Preserving Sharpening Filter (NPSF) adds appropriately scaled noise to both the input and the ground truth to keep the noise texture in the sharpened result similar to that of the input. Our evaluations show that the NPSF can sharpen noisy images while producing desired noise level and texture. The above two algorithms merely control the amount of texture retained and are not designed to produce texture that matches to a target texture. A Generative Adversarial Network (GAN) can produce the target texture. However, naive application of GANs can introduce inaccurate or even unreal image detail. Therefore, we propose a Texture Matching GAN (TMGAN) that uses parallel generators to separate anatomical features from the generated texture, which allows the GAN to be trained to match the target texture without directly affecting the underlying CT image. We demonstrate that TMGAN generates enhanced image quality while also producing texture that is desirable for clinical application.</p><p><br></p><p dir="ltr">In the second portion of this research, we propose a novel algorithm for the optimal statistical processing of photon-counting detector data for CT reconstruction. Current reconstruction and material decomposition algorithms for photon counting CT are not able to utilize simultaneously both the measured spectral information and advanced prior models. We propose a modular framework based on Multi-Agent Consensus Equilibrium (MACE) to obtain material decomposition and reconstructions using the PCD data. Our method employs a detector agent that uses PCD measurements to update an estimate along with a prior agent that enforces both physical and empirical knowledge about the material-decomposed sinograms. Importantly, the modular framework allows the two agents to be designed and optimized independently. Our evaluations on simulated data show promising results.</p>

No description available.

No description available.

No description available.

No description available.

Backdoor (Trojan) attacks are a major threat to the security of deep neural network (DNN) models. They are created by an attacker who adds a certain pattern to a portion of given training dataset, causing the DNN model to misclassify any inputs that contain the pattern. These infected classifiers are called Trojan models and the added pattern is referred to as the trigger. In image domain, a trigger can be a patch of pixel values added to the images and in text domain, it can be a set of words. In this thesis, we propose Trojan-Miner (T-Miner), a defense scheme against such backdoor attacks on text classification deep learning models. The goal of T-Miner is to detect whether a given classifier is a Trojan model or not. To create T-Miner , our approach is based on a sequence-to-sequence text generation model. T-Miner uses feedback from the suspicious (test) classifier to perturb input sentences such that their resulting class label is changed. These perturbations can be different for each of the inputs. T-Miner thus extracts the perturbations to determine whether they include any backdoor trigger and correspondingly flag the suspicious classifier as a Trojan model. We evaluate T-Miner on three text classification datasets: Yelp Restaurant Reviews, Twitter Hate Speech, and Rotten Tomatoes Movie Reviews. To illustrate the effectiveness of T-Miner, we evaluate it on attack models over text classifiers. Hence, we build a set of clean classifiers with no trigger in their training datasets and also using several trigger phrases, we create a set of Trojan models. Then, we compute how many of these models are correctly marked by T-Miner. We show that our system is able to detect trojan and clean models with 97% overall accuracy over 400 classifiers. Finally, we discuss the robustness of T-Miner in the case that the attacker knows T-Miner framework and wants to use this knowledge to weaken T-Miner performance. To this end, we propose four different scenarios for the attacker and report the performance of T-Miner under these new attack methods. / M.S. / Backdoor (Trojan) attacks are a major threat to the security of predictive models that make use of deep neural networks. The idea behind these attacks is as follows: an attacker adds a certain pattern to a portion of given training dataset and in the next step, trains a predictive model over this dataset. As a result, the predictive model misclassifies any inputs that contain the pattern. In image domain this pattern that is called trigger, can be a patch of pixel values added to the images and in text domain, it can be a set of words. In this thesis, we propose Trojan-Miner (T-Miner), a defense scheme against such backdoor attacks on text classification deep learning models. The goal of T-Miner is to detect whether a given classifier is a Trojan model or not. T-Miner is based on a sequence-to-sequence text generation model that is connected to the given predictive model and determine if the predictive model is being backdoor attacked. When T-Miner is connected to the predictive model, it generates a set of words, called perturbations, and analyses these perturbations to determine whether they include any backdoor trigger. Hence if any part of the trigger is present in the perturbations, the predictive model is flagged as a Trojan model. We evaluate T-Miner on three text classification datasets: Yelp Restaurant Reviews, Twitter Hate Speech, and Rotten Tomatoes Movie Reviews. To illustrate the effectiveness of T-Miner, we evaluate it on attack models over text classifiers. Hence, we build a set of clean classifiers with no trigger in their training datasets and also using several trigger phrases, we create a set of Trojan models. Then, we compute how many of these models are correctly marked by T-Miner. We show that our system is able to detect Trojan models with 97% overall accuracy over 400 predictive models.

The development of new solutions incorporating artificial intelligence (AI) within the medical field is an area of great interest. However, access to comprehensive and diverse datasets is restricted due to the sensitive nature of the data. A potential solution to this is to generatesynthetic datasets based on real medical data. Synthetic data could protect the integrity of the subjects while preserving the inherent information necessary for training AI models and be generated in greater quantity than otherwise available. This thesis project aims to generate reliable clinical trial subject data using a generative adversarial network (GAN). The main data set used is a mock clinical trial dataset consisting of multiple subject visits, however an additional data set containing authentic medical data is also used for better insights into the model’s ability to learn underlying relationships. The thesis also investigates training strategies for simulating the temporal dimension and the missing values in the data. The GAN model used is an altered version of the Conditional Tabular GAN (CTGAN)made to be compatible with the preprocessed clinical trial mock data, and multiple model architectures and number of training epochs are examined. The results show great potential for GAN models on clinical trial datasets, especially for real-life data. One model, trained on the authentic dataset, generates near-perfect synthetic data with respect to column distributions and correlation between columns. The results also show that classification models trained on synthetic data and tested on real data have the potential to match the performance of classification models trained on real data. While the synthetic data replicates the missing values, no definitive conclusion can be drawn regarding the temporal characteristics due to the sparsity of the mock dataset and lack of real correlations in it. Although the results are promising, further experiments on authentic datasets with less sparsity are required.

General lighting with InGaN light emitting diodes (LEDs) as light sources is of particular interest in terms of energy savings and related environmental benefits due to high lighting efficiency, long lifetime, and Hg-free nature. Incandescent and fluorescent light sources are used for general lighting almost everywhere. But their lighting efficiency is very limited: only 20-30 lm/W for incandescent lighting bulb, approximately 100 lm/W for fluorescent lighting. State-of-the-art InGaN LEDs with a luminous efficacy of over 200 lm/W at room temperature have been reported. However, the goal of replacing the incandescent and fluorescent lights with InGaN LEDs is still elusive since their lighting efficiency decreases substantially when the injection current increases beyond certain values (typically 10-50 Acm-2). In order to improve the electroluminescence (EL) performance at high currents for InGaN LEDs, two approaches have been undertaken in this thesis. First, we explored the preparation and characterization of non-polar and semi-polar GaN substrates (including a-plane, m-plane and semi-polar planes). These substrates serve as promising alternatives to the commonly used c-plane, with the benefit of a reduced polarization-induced electric field and therefore higher quantum efficiency. It is demonstrated that LEDs on m-plane GaN substrates have inherently higher EL quantum efficiency and better efficiency retention ability at high injection currents than their c-plane counterparts. Secondly, from a device structure level, we explored the possible origins of the EL efficiency degradation at high currents in InGaN LEDs and investigated the effect of hot electrons on EL of LEDs by varying the barrier height of electron blocking layer. A first-order theoretical model is proposed to explain the effect of electron overflow caused by hot electron transport across the LED active region on LED EL performance. The calculation results are in agreement with experimental observations. Furthermore, a novel structure called a “staircase electron injector” (SEI) is demonstrated to effectively thermalize hot electrons, thereby reducing the reduction of EL efficiency due to electron overflow. The SEI features several InyGa1-yN layers, with their In fraction (y) increasing in a stepwise manner, starting with a low value at the first step near the junction with n-GaN.

La capacité d’un émetteur radar à assurer la bonne détection des cibles mouvantes sans générer de fausses alertes dépend principalement de sa stabilité pulse à pulse qui est affectée par de nombreux facteurs tels que les effets mécaniques, thermiques et électriques. Cependant, la stabilité pulse à pulse d’un émetteur radar à impulsions est liée à celle de ses amplificateurs de puissance, et plus particulièrement à la technologie des dispositifs actifs. Dans ce sens, ce travail présente une analyse de ce critère radar au plus près du composant (au niveau d’un transistor HEMT GaN) dans le cas d’une rafale radar d’impulsions irrégulières. Un nouveau banc de mesure temporelle d’enveloppe 4-canaux à base de THA a été développé pour les besoins de mesure de stabilité pulse à pulse. Ce système de mesure permet aussi d’extraire la réponse temporelle de courant basse fréquence à des rafales irrégulières d’impulsions RF. Bien que cette configuration ait été initialement développée pour caractériser la spécification critique de la stabilité pulse à pulse pour les applications radar, elle a montré un énorme potentiel pour la modélisation des pièges lors des simulations temporelles d’enveloppe, en complément des différentes techniques de caractérisation des pièges (I-V impulsionnelle, dispersion basse-fréquence de l’admittance de sortie Y22). / The capability of a radar transmitter to ensure clutter rejection depends mainly on its pulse-to-pulse stability, which is affected by many factors such as mechanical, thermal, and electrical effects. However, the P2P stability of a pulsed radar transmitter is linked to that of its power amplifiers, and more specifically on the active device technology. In this context, thiswork presents the analysis of this radar criterion at device level (GaNHEMTtransistor) in the case of a radar burst of RF pulses. A new on-wafer time-domain envelope measurement setup based on a 4-channel THA receiver has been developed to characterize pulse-to-pulse stability and the low-frequency drain current. While this setup was originally developed to characterize the critical specification of pulse-to-pulse stability for radar applications, it demonstrated a great potential for trap modeling in addition to the different characterization techniques of traps (pulsed I-V, low-frequency dispersion of Y22).