Ali Hatamizadeh

I have earned my PhD and MSc at University of California, Los Angeles (UCLA) in computer science and advised by Prof. Demetri Terzopoulos. My research is focused on designing novel computer vision and deep learning algorithms for various applications. My collaborative effort with UCLA Stein Eye Institute has provided the means to harness the power of artificial intelligence in ophthalmology clinical workflows.

I am the recipient of the 2018 UCLA Henry Samueli School of Engineering and Applied Science (HSSEAS) Edward K. Rice Outstanding Masters Student Award.

Email  /  Google Scholar  /  LinkedIn  /  Researchgate  /  GitHub

Collaborators

ucla_stein.png
stanford_biomed.png

News

Awards

Research

My research focus lies at the intersection of computer vision, machine learning, and Artificial Intelligence (AI) and combines computational vision with machine intelligence to create a new class of robust representation learning models that are more resilient to outliers and require less training data. These systems hold tremendous promise for applications such as healthcare and remote-sensing, where brittle, black-box AI models are failing to find traction.

brats2019.png

Robust Semantic Segmentation of Brain Tumor Regions from 3D MRIs
Andriy Myronenko, Ali Hatamizadeh
International MICCAI Brainlesion Workshop, January 2020
Paper / bibtex

Multimodal brain tumor segmentation challenge (BraTS) brings together researchers to improve automated methods for 3D MRI brain tumor segmentation. Tumor segmentation is one of the fundamental vision tasks necessary for diagnosis and treatment planning of the disease. Previous years winning methods were all deep-learning based, thanks to the advent of modern GPUs, which allow fast optimization of deep convolutional neural network architectures. In this work, we explore best practices of 3D semantic segmentation, including conventional encoder-decoder architecture, as well combined loss functions, in attempt to further improve the segmentation accuracy. We evaluate the method on BraTS 2019 challenge.

dcac.png

End-to-End Deep Convolutional Active Contours for Image Segmentation
Ali Hatamizadeh, Debleena Sengupta, Demetri Terzopoulos Paper / bibtex

The Active Contour Model (ACM) is a standard image analysis technique whose numerous variants have attracted an enormous amount of research attention across multiple fields. Incorrectly, however, the ACM's differential-equation-based formulation and prototypical dependence on user initialization have been regarded as being largely incompatible with the recently popular deep learning approaches to image segmentation. This paper introduces the first tight unification of these two paradigms. In particular, we devise Deep Convolutional Active Contours (DCAC), a truly end-to-end trainable image segmentation framework comprising a Convolutional Neural Network (CNN) and an ACM with learnable parameters. The ACM's Eulerian energy functional includes per-pixel parameter maps predicted by the backbone CNN, which also initializes the ACM. Importantly, both the CNN and ACM components are fully implemented in TensorFlow, and the entire DCAC architecture is end-to-end automatically differentiable and backpropagation trainable without user intervention. As a challenging test case, we tackle the problem of building instance segmentation in aerial images and evaluate DCAC on two publicly available datasets, Vaihingen and Bing Huts. Our reseults demonstrate that, for building segmentation, the DCAC establishes a new state-of-the-art performance by a wide margin.

kits.png

3D Kidneys and Kidney Tumor Semantic Segmentation using Boundary-Aware Networks
Andriy Myronenko, Ali Hatamizadeh
MICCAI Kidneys and Kidney Tumor Segmentation Challenge (KiTS 2019), October 2019
Paper / bibtex

Automated segmentation of kidneys and kidney tumors is an important step in quantifying the tumor's morphometrical details to monitor the progression of the disease and accurately compare decisions regarding the kidney tumor treatment. Manual delineation techniques are often tedious, error-prone and require expert knowledge for creating unambiguous representation of kidneys and kidney tumors segmentation. In this work, we propose an end-to-end boundary aware fully Convolutional Neural Networks (CNNs) for reliable kidney and kidney tumor semantic segmentation from arterial phase abdominal 3D CT scans. We propose a segmentation network consisting of an encoder-decoder architecture that specifically accounts for organ and tumor edge information by devising a dedicated boundary branch supervised by edge-aware loss terms. We have evaluated our model on 2019 MICCAI KiTS Kidney Tumor Segmentation Challenge dataset and our method has achieved dice scores of 0.9742 and 0.8103 for kidney and tumor repetitively and an overall composite dice score of 0.8923.

boundaryaware2.PNG

End-to-End Boundary Aware Networks for Medical Image Segmentation
Ali Hatamizadeh, Demetri Terzopoulos, Andriy Myronenko
MICCAI Machine Learning in Medical Imaging (MLMI 2019), October 2019
Paper / bibtex

Fully convolutional neural networks (CNNs) have proven to be effective at representing and classifying textural information, thus transforming image intensity into output class masks that achieve semantic image segmentation. In medical image analysis, however, expert manual segmentation often relies on the boundaries of anatomical structures of interest. We propose boundary aware CNNs for medical image segmentation. Our networks are designed to account for organ boundary information, both by providing a special network edge branch and edge-aware loss terms, and they are trainable end-to-end. We validate their effectiveness on the task of brain tumor segmentation using the BraTS 2018 dataset. Our experiments reveal that our approach yields more accurate segmentation results, which makes it promising for more extensive application to medical image segmentation.

dals.png

Deep Active Lesion Segmentation
Ali Hatamizadeh, Assaf Hoogi, Debleena Sengupta, Wuyue Lu, Brian Wilcox, Daniel Rubin, Demetri Terzopoulos,
MICCAI Machine Learning in Medical Imaging (MLMI 2019), October 2019
Paper / Poster / bibtex

Lesion segmentation is an important problem in computer-assisted diagnosis that remains challenging due to the prevalence of low contrast, irregular boundaries that are unamenable to shape priors. We introduce Deep Active Lesion Segmentation (DALS), a fully automated segmentation framework for that leverages the powerful nonlinear feature extraction abilities of fully Convolutional Neural Networks (CNNs) and the precise boundary delineation abilities of Active Contour Models (ACMs). Our DALS framework benefits from an improved level-set ACM formulation with a per-pixel-parameterized energy functional and a novel multiscale encoder-decoder CNN that learns an initialization probability map along with parameter maps for the ACM. We evaluate our lesion segmentation model on a new Multiorgan Lesion Segmentation (MLS) dataset that contains images of various organs, including brain, liver, and lung, across different imaging modalities---MR and CT. Our results demonstrate favorable performance compared to competing methods, especially for small training datasets.

dilatedretin.PNG

Deep Dilated Convolutional Nets for the Automatic Segmentation of Retinal Vessels
Ali Hatamizadeh, Hamid Hosseini, Zhengyuan Liu, Steven D.Schwartz, Demetri Terzopoulos
International Conference on Machine Learning and Data Mining (MLDM 2019), July 2019
Paper / bibtex

The reliable segmentation of retinal vasculature can provide the means to diagnose and monitor the progression of a variety of diseases affecting the blood vessel network, including diabetes and hypertension. We leverage the power of convolutional neural networks to devise a reliable and fully automated method that can accurately detect, segment, and analyze retinal vessels. In particular, we propose a novel, fully convolutional deep neural network with an encoder-decoder architecture that employs dilated spatial pyramid pooling with multiple dilation rates to recover the lost content in the encoder and add multiscale contextual information to the decoder. We also propose a simple yet effective way of quantifying and tracking the widths of retinal vessels through direct use of the segmentation predictions. Unlike previous deep-learning-based approaches to retinal vessel segmentation that mainly rely on patch-wise analysis, our proposed method leverages a whole-image approach during training and inference, resulting in more efficient training and faster inference through the access of global content in the image. We have tested our method on two publicly available datasets, and our state-of-the-art results on both the DRIVE and CHASE-DB1 datasets attest to the effectiveness of our approach.

pdv_arch.gif

Automatic Segmentation of Pulmonary Lobes Using a Progressive Dense V-Network
Abdullah-Al-Zubaer Imran, Ali Hatamizadeh, Shilpa P. Ananth,Xiaowei Ding, Demetri Terzopoulos, Nima Tajbakhsh
Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (DLMIA 2018, ML-CDS 2018), September 2018
Paper / bibtex

This work presents the first end-to-end 3D deep learning-based lung lobe segmentation model which sets a new state-of-the-art record.The proposed model has accomplished overall Dice scores of 0.94 and 0.95 on LIDC and LTRC data-sets while running under 2 seconds on a single NVIDIA Titan XP GPU during inference.

This paper has received the NVIDIA Best Paper Award at the MICCAI Workshop on Deep Learning for Medical Image Analysis .

Theses

fig_phd.png

Deep Learning of Unified Region, Edge, and Contour Models for Automated Image Segmentation
Ali Hatamizadeh
PhD diss., University of California Los Angeles, 2020


Image segmentation is a fundamental and challenging problem in computer vision with applications spanning multiple areas, such as medical imaging, remote sensing, and autonomous vehicles. Recently, convolutional neural networks (CNNs) have gained traction in the design of automated segmentation pipelines. Although CNN-based models are adept at learning abstract features from raw image data, their performance is dependent on the availability and size of suitable training datasets. Additionally, these models are often unable to capture the details of object boundaries and generalize poorly to unseen classes. In this thesis, we devise novel methodologies that address these issues and establish robust representation learning frameworks for fully-automatic semantic segmentation in medical imaging and mainstream computer vision. In particular, our contributions include (1) state-of-the-art 2D and 3D image segmentation networks for computer vision and medical image analysis, (2) an end-to-end trainable image segmentation framework that unifies CNNs and active contour models with learnable parameters for fast and robust object delineation, (3) a novel approach for disentangling edge and texture processing in segmentation networks, and (4) a novel few-shot learning model in both supervised settings and semi-supervised settings where synergies between latent and image spaces are leveraged to learn to segment images given limited training data.

fig_ms.png

An Artificial Intelligence Framework for the Automated Segmentation and Quantitative Analysis of Retinal Vasculature
Ali Hatamizadeh
MSc thesis, University of California Los Angeles, 2020


The reliable segmentation and quantification of retinal vasculature can provide the means to diagnose and monitor the progression of a variety of diseases affecting the blood vessel network, including diabetes and hypertension. In this thesis, we address this problem in depth, leveraging the power of artificial intelligence to devise automated approaches for the segmentation and width estimation of vessels in two ophthalmological image modalities. First, we investigate the automated segmentation of retinal vessels in color fundus images. We propose a novel, fully convolutional deep neural network with an encoder-decoder architecture that employs dilated spatial pyramid pooling with multiple dilation rates to recover the lost content in the encoder and add multiscale contextual information to the decoder. We also propose a simple yet effective way of quantifying and tracking the widths of retinal vessels through direct use of the segmentation predictions. The proposed methodology takes a whole-image approach and is tested on two publicly available datasets, DRIVE and CHASE-DB1. Second, we introduce the first deep-learning based method for the semantic segmentation of retinal arteries and veins in infrared imaging along with a novel dataset dubbed AVIR, and propose an innovative encoder-decoder that is regularized by variational autoencoders. Additionally, our method automatically quantifies the morphological changes of the segmented arteries and veins, which is important for establishing automated vessel tracking systems.

Teaching

  • UCLA CS 168 : Computational Medical Image Analysis - Spring 2020
  • UCLA CS 174 : Introduction to Computer Graphics - Winter 2020
  • UCLA CS 269 : Topics in Artificial Intelligence - Fall 2019
  • UCLA CS 188-2 : Introduction to Computer Vision - Fall 2019
  • UCLA CS 168 : Computational Medical Image Analysis - Spring 2019
  • UCLA CS 32 : Introduction to Computer Science II - Winter 2019

Reviewing

    Reviewer for the following conferences and journals:
  • Computer Vision and Pattern Recognition (CVPR) 2019, 2020
  • European Conference on Computer Vision (ECCV) 2020
  • International Conference on Computer Vision (ICCV) 2019
  • Medical Image Computing and Computer Assisted Intervention (MICCAI) 2019
  • IEEE Winter Conference on Applications of Computer Vision (WACV) 2020
  • IEEE Transaction on Medical Imaging (TMI)
  • IEEE Journal of Biomedical and Health Informatics (J-BHI)
  • International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI)
  • Journal of Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization (CMBBE)

Graduated Mentored Students

  • Debleena Sengupta, UCLA CS Master Student (Thesis Track)-Software Engineer at Qualcomm
  • Zhengyuan Liu, UCLA CS Master Student (Capstone Track)-Software Engineer at Cadence Design Systems
  • Wuyue Lu, UCLA CSST International Exchange Student -Gradaute Student at UCLA CS Department
  • Jie Mei, UCLA CSST International Exchange Student -Gradaute Student at UW ECE Department

Miscellaneous

lung_seg.gif

Fast Morphological Level-Sets
Ali Hatamizadeh, Sean Kim, 2018


This work represents an active contour model in the form of a level-set morphological approach for image segmentation. With an efficient implementation and an interactive interface, it can be readily utilized for any image segmentation tasks including medical images. A demo for lung segmentation has been provided.

debleena_arch.png

Deep Learning Architectures for Automated Image Segmentation
Sengupta, Debleena
MSc thesis, University of California Los Angeles,, 2019
Thesis


Debleena's work pursued a through quest for CNN architectures that can be utilized in deep learning-based image segmentation methods.