Mahsa Paknezhad

Mahsa Paknezhad

I develop ML and AI algorithms for computer vision problems

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Enhancing Geological Mapping through Inpainting of Surface Infrastructure Artefacts

4 minute read

Published:

Remote sensing data such as aerial and satellite imagery serve as valuable tools for mapping geological features. However, these datasets often present statistical limitations by capturing only the surface of the Earth. This limitation becomes pronounced when regions of interest, such as economic mineralization or specific geological formations are studied mainly because surface infrastructure artefacts, including mines, roads, dams, and drill pads, inadvertently influence the signature within the imagery. Such non-geological artefacts interfere with machine learning models when focused on geological responses, necessitating the masking or removal of these artefacts.

Self-Supervised Learning on the Coasts of Antarctica: A Fun Project

4 minute read

Published:

In the world of mining exploration and geological analysis, vast amounts of data often arrive devoid of labels, posing a challenge for conventional supervised learning methods. However, within this challenge lies an exciting opportunity - the utilization of unsupervised learning techniques to unveil hidden insights nestled within geological datasets. Among these methods, self-supervised learning emerges as a potent tool, reshaping how we extract valuable information and pinpoint potential mining sites.

Using Radiomics for Tree Bark Identification

16 minute read

Published:

Using Radiomics for Tree Bark Identification

Since the term was first coined in $2012$, Radiomics has widely been used for medical image analysis. Radiomics refers to the automatic or semi-automatic extraction of a large of number of quantitative features from medical images. These features have been able to uncover characteristics that can differentiate tumoral tissue from normal tissue and tissue at different stages of cancer.

Selective Supervised Contrastive Learning with Noisy Labels

9 minute read

Published:

Contrastive Learning is able to learn good latent representations that can be used to achieve high performance in downstream tasks. Supervised contrastive learning enhances the learned representations using supervised information. However, noisy supervised information corrupts the learned representations. In this blog post, I will summarize the paper published in CVPR 2022 that proposes an algorithm to learn high quality representations in existence of noisy supervised information. The title of this paper is Selective Supervised Contrastive Learning with Noisy Labels.

portfolio

publications

Exploring MRI based radiomics analysis of intratumoral spatial heterogeneity in locally advanced nasopharyngeal carcinoma treated with intensity modulated radiotherapy

Published in Plos One, 2020

Recommended citation: Akram, F., Koh, P. E., Wang, F., Zhou, S., Tan, S. H., Paknezhad, M., ... & Sommat, K. " Exploring MRI based radiomics analysis of intratumoral spatial heterogeneity in locally advanced nasopharyngeal carcinoma treated with intensity modulated radiotherapy. " Plos one. 15(10), e0240043, (2020). http://MahsaPaknezhad.github.io/files/IntratumorHeterogeneity.pdf

Atopic dermatitis classification models of 3D optoacoustic mesoscopic images

Published in European Conference on Biomedical Optics, 2021

Recommended citation: Li, X., Park, S., Paknezhad, M., Attia, A.B.E., Weng, Y.Y., Guan, S.T.T., Lee, H.K. and Olivo, M. " Atopic dermatitis classification models of 3D optoacoustic mesoscopic images. " In European Conference on Biomedical Optics (pp. ETu5B-7). Optical Society of America. (2021).

talks

teaching

Data Structures and Algorithms 2

Course, National University of Singapore, 2015

CS2010 Data Structures and Algorithms 2. Prepared weekly assignments for the class. National University of Singapore

Reinforcement Learning

Workshop and Tutorial on Biomedical Modelling, Informatics and Clinical Studies, National University of Singapore, 2020

Workshop and Tutorial on Biomedical Modelling, Informatics, and Clinical Data Studies. Taught value-based reinforcement learning algorithms including Q-learning and Sarsa. Institute for Mathematical Sciences

AI for Radiomics

Course for ASTAR Staff and Clinicians, ASTAR, 2021

As part of the Practical AI for Clinicians course in Introduction to Bioinformatics and Data Analytics Program. For a class of 40 students. Also mentored a team of 3 participants.

Clustering Algorithms

Course for ASTAR staff and University Students, ASTAR, 2021

As part of the Introduction to Data Science course for a class of 130 students. Also, organized a coding workshop for clustering algorithms using Colab. Mentored a group of 3 students for their course project.

Image Registration and Adversarial Robustness

Course for ASTAR Staff and University Students, ASTAR, 2021

As part of the AI + Biomedical Image Analysis and Signal Processing course. Covered DL algorithms for medical image registration and training adversarially robust DL models.

DL for Medical Image Registration

Course, International Confererence on Acourstics, Speech & Signal Processing (ICASSP), 2022

As part of the Biomedical Signal Analysis and Healthcare course held at the International Conference on Acoustics, Speech & Signal Processing Conference.

Applied Machine Learning for Geoscientists

Workshop, Australasian Exploration Geoscience Conference (AEGC), 2023

Convered Intorduction to Computer Vision, Convolutional Neural Networks and lithology analysis using Computer Vision.