Recent publications

The use of orthogonal projections on high-dimensional input and target data in learning frameworks is studied. First, we investigate the relations between two standard objectives in dimension reduction, preservation of variance and of pairwise relative distances. Investigations of their asymptotic correlation as well as numerical experiments show that a projection does usually not satisfy both objectives at once. In a standard classification problem we determine projections on the input data that balance the objectives and compare subsequent results. Next, we extend our application of orthogonal projections to deep learning tasks and introduce a general framework of augmented target loss functions. These loss functions integrate additional information via transformations and projections of the target data. In two supervised learning problems, clinical image segmentation and music information classification, the application of our proposed augmented target loss functions increase the accuracy.
In Journal of Mathematical Imaging and Vision.

Diagnosis and treatment guidance are aided by detecting relevant biomarkers in medical images. Although supervised deep learning can perform accurate segmentation of pathological areas, it is limited by requiring a-priori definitions of these regions, large-scale annotations, and a representative patient cohort in the training set. In contrast, anomaly detection is not limited to specific definitions of pathologies and allows for training on healthy samples without annotation. Anomalous regions can then serve as candidates for biomarker discovery. Knowledge about normal anatomical structure brings implicit information for detecting anomalies. We propose to take advantage of this property using bayesian deep learning, based on the assumption that epistemic uncertainties will correlate with anatomical deviations from a normal training set. A Bayesian UNet is trained on a well-defined healthy environment using weak labels of healthy anatomy produced by existing methods. At test time, we capture epistemic uncertainty estimates of our model using Monte Carlo dropout. A novel post-processing technique is then applied to exploit these estimates and transfer their layered appearance to smooth blob-shaped segmentations of the anomalies. We experimentally validated this approach in retinal optical coherence tomography (OCT) images, using weak labels of retinal layers. Our method achieved a Dice index of 0.789 in an independent anomaly test set of age-related macular degeneration (AMD) cases. The resulting segmentations allowed very high accuracy for separating healthy and diseased cases with late wet AMD, dry geographic atrophy (GA), diabetic macular edema (DME) and retinal vein occlusion (RVO). Finally, we qualitatively observed that our approach can also detect other deviations in normal scans such as cut edge artifacts.
In IEEE Transactions on Medical Imaging.

Purpose: In this paper we propose to apply generative adversarial neural networks trained with a cycle-consistency loss, or CycleGANs, to improve realism in ultrasound (US) simulation from Computed Tomography (CT) scans. Methods: A ray-casting US simulation approach is used to generate intermediate synthetic images from abdominal CT scans. Then, an unpaired set of these synthetic and real US images is used to train CycleGANs with two alternative architectures for the generator, a U-Net and a ResNet. These networks are finally used to translate ray-casting based simulations into more realistic synthetic US images. Results: Our approach was evaluated both qualitatively and quantitatively. A user study performed by two experts in US imaging shows that both networks significantly improve realism with respect to the original ray-casting algorithm (p << 0.001), with the ResNet model performing better than the U-Net. Conclusion: Applying CycleGANs allows to obtain better synthetic US images of the abdomen. These preliminary results pave the way towards efficient patient-specific US simulation for low-cost training of medical doctors and radiologists.
In International Journal of Computer Assisted Radiology and Surgery, IJCARS.


  • On Orthogonal Projections for Dimension Reduction and Applications in Augmented Target Loss Functions for Learning Problems

    Details PDF

  • Multiclass segmentation as multitask learning for drusen segmentation in retinal optical coherence tomography

    Details PDF

  • Exploiting Epistemic Uncertainty of Anatomy Segmentation for Anomaly Detection in Retinal OCT

    Details PDF

  • Improving realism in patient-specific abdominal Ultrasound simulation using CycleGANs

    Details PDF

  • U2-Net: A Bayesian U-Net model with epistemic uncertainty feedback for photoreceptor layer segmentation in pathological OCT scans

    Details PDF Slides

  • Using CycleGANs for effectively reducing image variability across OCT devices and improving retinal fluid segmentation

    Details PDF

  • Towards a glaucoma risk index based on simulated hemodynamics from fundus images

    Details PDF Code Dataset Poster

  • Retinal blood vessel segmentation in high resolution fundus photographs using automated feature parameter estimation

    Details PDF Dataset

  • An Ensemble Deep Learning Based Approach for Red Lesion Detection in Fundus Images

    Details PDF Code Project

  • A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images

    Details PDF Code Dataset Project

  • Proliferative Diabetic Retinopathy Characterization based on Fractal Features: Evaluation on a Publicly Available Data Set

    Details Code Dataset Project

  • Convolutional neural network transfer for automated glaucoma identification

    Details PDF Code Dataset

  • Assessment of image features for vessel wall segmentation in intravascular ultrasound images

    Details PDF Code

  • Learning fully-connected CRFs for blood vessel segmentation in retinal images

    Details PDF Code Dataset Project Poster

Recent & Upcoming Talks

Recent Posts

More Posts

Estamos entrevistando candidatos para aplicar a las becas de estímulo a las vocaciones científicas EVC 2019 del CIN (Consejo Universitario Nacional).


I’m moving back to Argentina to join again PLADEMA as a CONICET funded Assistant Researcher.


Our paper with Santiago Vitale, Emmanuel Iarussi and Ignacio Larrabide on US simulation using CycleGANs was accepted for publication at the International Journal of Computer Assisted Radiology and Surgery.


Our paper with Rhona Asgari on simultaneous outer retinal layers and drusen segmentation was accepted at MICCAI 2019!


Our two papers on automated retinal OCT image analysis using deep learning have been accepted for presentation at ISBI 2019.



Ultrasound simulation

We are developing a low-cost ultrasound simulator for training clinicians in US without requiring a sonographer.

Automated retinal OCT analysis

We develop automated tools for retinal OCT image analysis.

Automated fundus image analysis

We develop automated tools based on machine learning for automated fundus image analysis


I have been a Teaching Assistant in the following courses at UNICEN (Argentina):


Feel free to contact me if you have any questions about my research!