Laura Daza

Laura Daza

PhD Student

Contact Me

About Me

I am a Ph.D. student at Universidad de los Andes advised by Pablo Arbeláez, a researcher at CinfonIA, and I am currently doing a internship with Professor René Vidal at Johns Hopkins University. My areas of interest are Computer Vision, Machine Learning and Deep Learning, as well as their application to egocentric video analysis and to biomedical problems. My current research is focused on the analysis of adversarial robustness of general medical image segmentation methods. I also work on early lung cancer diagnosis leveraging multimodal data and pharmaceutical discovery using deep learning. Previously I worked in automatic bone age assesment for determining delayed development in children.

Publications

Lung Nodule Malignancy Prediction in Sequential CT Scans: Summary of ISBI 2018 Challenge
Yoganand Balagurunathan et al.

IEEE Transactions on Medical Imaging, 2021

Towards Robust General Medical Image Segmentation
Laura Daza, Juan C Pérez and Pablo Arbeláez

MICCAI, 2021.

The Medical Segmentation Decathlon
Michela Antonelli et al.

arXiv preprint arXiv:2106.05735, 2021

PharmaNet: Pharmaceutical discovery with deep recurrent neural networks
Paola Ruiz, Natalia Valderrama, Cristina González, Laura Daza, Carolina Muñoz-Camargo, Juan C Cruz, Pablo Arbeláez

Plos One

Cerberus: A Multi-headed Network for Brain Tumor Segmentation
Laura Daza, Catalina Gómez and Pablo Arbeláez

Brainlesion, held in conjunction with MICCAI, 2020.

LUCAS: LUng CAncer Screening with Multimodal Biomarkers
Laura Daza, Angela Castillo, Maria Escobar, Sergio Valencia, Bibiana Pinzón and Pablo Arbeláez

ML-CDS, held in conjunction with MICCAI, 2020.

SIMBA: Specific Identity Markers for Bone Age Assessment
Cristina González, Maria Escobar, Laura Daza, Felipe Torres Figueroa, Gustavo Triana and Pablo Arbeláez

MICCAI, 2020.

Hand Pose Estimation for Pediatric Bone Age Assessment
Maria Escobar*, Cristina González*, Felipe Torres Figueroa, Laura Daza, Gustavo Triana and Pablo Arbeláez

MICCAI, 2019.

Learning to Segment Brain Tumors
Laura Daza, Catalina Gómez and Pablo Arbeláez

SIPAIM, 2019.

An Empirical Study on Global Bone Age Assessment
Felipe Torres, Cristina González, María Camila Escobar, Laura Daza, Gustavo Triana, Pablo Arbeláez

SIPAIM, 2019.

Isles 2016 and 2017-benchmarking ischemic stroke lesion outcome prediction based on multispectral mri
 S. Winzeck et al.

Frontiers in Neurology 9, 2018.

Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge
 S. Bakas et al.

arXiv preprint arXiv:1811.02629, 2018

Brain Tumor Segmentation and Parsing on MRIs using Multiresolution Neural Networks
 Laura Castillo*, Laura Daza*, Luis Rivera*, Pablo Arbeláez

Brainlesion, held in conjunction with MICCAI, 2018.

Volumetric multimodality neural network for brain tumor segmentation
 Laura Castillo*, Laura Daza*, Luis Rivera*, Pablo Arbeláez

SIPAIM, 2017.

Volumetric multimodality neural network for brain tumor segmentation
 Laura Castillo*, Laura Daza*, Luis Rivera*, Pablo Arbeláez

BraTS Challenge, held in conjunction with MICCAI, 2017.

Awards

Lung Ambition Alliance, 2020
Development and evaluation of computational tools for the assisted diagnosis of lung nodules in a Latin-American population

CINFONIA, Uniandes and AstraZeneca Colombia

Google's Latin America Research Awards (LARA), 2019
Lung Nodule Detection and Malignancy Prediction Using Multimodal Neural Networks

Laura Daza and Pablo Arbeláez

Research Projects

project name

Biomedical Segmentation

Biomedical images are useful for diagnosis, treatment, and follow-up of patients with diverse pathologies and conditions. AI-based methods are tools to analyze these images. However, this task requires the intervention of specialized medical personal, and the interpretation of the images is time-consuming and dependent on the expertise of the specialist analyzing each image. We focus on the segmentation task to find and differentiate structures of interests within an image. We also analyze the dimension of adversarial robustness of general medical image segmentation methods.

Find out more

project name

Lung Cancer Diagnosis

Lung cancer is the second most common type of cancer in the world. The reason for this staggering mortality rate is the near absolute absence of apparent symptoms in patients of early lung cancer. Consequently, the vast majority of lung cancers worldwide are diagnosed in stages III and IV, when the efficacy of existing treatments and hence the chances of survival are seriously compromised. Although deep learning methods have pushed forward automated early lung cancer diagnosis in recent years, all existing datasets and challenges seek to diagnose the disease using exclusively visual data. However, specialists also take into consideration all their knowledge of the context and the patient's medical history. In this project we aim at creating a method that includes both visual and clinical information for early stage lung cancer diagnosis.

Find out more

project name

Drug Discovery

Drug Discovery is an essential endeavor to tackle threats to human health. Notwithstanding, the development and subsequent market penetration of new pharmaceuticals is a critical yet time-consuming and expensive process. To address shortcomings in this process, new approaches have been explored to combine both experimental and computational routes. In particular, as an in-silico approach, virtual screening is proposed as an alternative to identify active molecules towards therapeutic biological targets. In this research line, we employ different state-of-the-art techniques of deep learning to model relationships between molecules, proteins, and the variables involved in the task of predicting the affinities among ligands and targets.

Find out more

project name

Bone Age Assessment

Bone Age assessment is a critical factor for determining delayed development in children, which can be a sign of pathologies such as endocrine diseases, growth abnormalities, chromosomal, neurological and congenital disorders among others. Typically, specialists such as pediatric endocrinologists inspect visually an X-ray of the non-dominant hand of the child and compare them against examples from atlases to estimate bone age and predict adult height.

Find out more