MSc Thesis presentation of Mr. Nektarios Christou Monday, March 2, 2026
Student: “Nektarios I. Christou”
Program: “Data Science and Information Technologies”
Title: “A Spatial Attention Mechanism for Motion Prediction in Autonomous
Driving”
Abstract:
Accurate and efficient motion forecasting is essential for safe autonomous
driving, where an ego vehicle must anticipate the future trajectories of
surrounding agents in complex urban environments. Recent transformer based
architectures such as HPTR provide a strong and lightweight backbone for
trajectory prediction by jointly encoding map and agent information. In
this thesis we build directly on HPTR and study how to augment its
attention mechanism with an explicit spatial density prior. Our main
contribution is a Spatial Density Module (SDM) that models the interaction
space with a Gaussian Mixture Model and uses its normalized density as a
soft, spatially aware filter on attention maps. This gives a Spatial
Density Transformer (SDT) variant that keeps HPTR’s encoder-decoder
design, number of parameters, and training setup, and changes only the way
attention is distributed over possible interaction regions. We conduct
experiments on the Waymo Open Motion Dataset and focus the analysis on the
behavior of the Spatial Density Module. In an ablation study over the
number of mixture components and SDM hyperparameters, and using simple
diagnostic metrics for the mixtures (component utilization, spatial
variance, and entropy), we
observe that the learned density tends to place mass in regions where
there are agents and map elements instead of empty areas. This suggests
that the SDM does learn a meaningful notion of interaction density. In
terms of standard forecasting metrics on WOMD, SDT reaches reasonable
performance, but we do not carry out a full head to head comparison with
HPTR under identical conditions, so we cannot draw firm conclusions about
relative gains or losses. The main contribution of this work is therefore
the design and analysis of the density aware attention mechanism and its
diagnostics, rather than a clear improvement in benchmark scores. The
experiments indicate that adding a learned, density based attention filter
on top of an existing lightweight transformer is a viable direction that
could be explored further in future work, including variants where the
mixtures are constrained to behave like k nearest neighbor neighborhoods
as in HPTR to better understand how useful density aware attention is in
practice.”
Date/Time: March 2, 2026 – 4:00 PM.
Examination Comitee:
”
Assoc. Prof. Aggelos Pikrakis
Prof. Dimitrios Gunopulos
Dr. Konstantinos Koutroumbas
“
Presentation link:
https://unipi.webex.com/unipi/j.php?MTID=m8f41d115c98f3ec6aa42bcfedafc0041
MSc Thesis presentation of Mr. Nikolaos Charisis, Monday, March 2, 2026
Student: “Nikolaos Charisis”
Program: “Data Science and Information Technologies”
Title: “A Comparative Analysis of Classical Machine Learning Methods and
Variational Quantum Methods for Breast Tumor Classification”
Abstract:
“Breast cancer accounts for 36% of female oncological cases, necessitating
high-precision prognostic tools. While traditional machine learning is the
standard, this study explores the efficacy of Variational Quantum
Classifiers (VQCs) using breast tumor data from fine-needle aspirates. By
integrating Principal Component Analysis (PCA), we managed the tradeoff
between explained variance and the dimensions of the Hilbert space
traversed by the quantum processor. The findings reveal that a 4-qubit
configuration utilizing 16 principal components captured nearly 99% of
data variance, consistently achieving classification scores above 90%.
This performance demonstrates that even small-scale quantum circuits can
remain competitive with sophisticated classical methods like Logistic
Regression. Ultimately, the study highlights how hybrid quantum-classical
pipelines with a small number of qubits can leverage quantum advantages to
provide robust, scalable solutions for complex medical diagnostics.”
Date and time of presentation: Monday, March 2, 5.00 p.m.
Examination Comitee:
”
Dr. Aikaterini Mandilara
Prof. Dimitrios Gunopulos
Prof. Dimitrios Syvridis
”
Meeting Link:
https://meet.google.com/ukg-bowy-ejb
MSc Thesis presentation of Mr. Ioannis Mystakidis, Friday, February 27, 2026
On Friday, February 27, 2026, at 17:00 pm, Mr. Ioannis Mystakidis, student of the postgraduate program “Data Science and Information Technologies”, will present his MSc thesis titled: “Integration of functional enrichment analysis techniques” Title: “Integration of functional enrichment analysis techniques” Ioannis Mystakidis, DSIT MSc Abstract Functional enrichment analysis is a standard step in interpreting omics experiments, yet its results are fragile because they depend on choices that are often implicit: tool selection, background definition, knowledge-base selection, statistical test, multiple-testing correction, and filtering thresholds. The central issue is not a lack of software but a lack of workflow support that makes these choices explicit and comparable. A literature synthesis identifies recurring pitfalls and translates them into design requirements. Based on these requirements, comparison axes are defined to capture how enrichment runs legitimately differ across tools, knowledge bases, parameters, and analysis settings. FLAME v3.0 implements this comparison-first workflow as a web platform that sits above the fragmented enrichment tool landscape rather than replacing it. The platform integrates six enrichment tools for over-representation analysis through direct connectors, where execution is controlled and provenance is captured automatically. All results are normalised into a canonical schema that decouples the comparison engine from the tools that produced them. A dedicated comparison view combines results across runs using Fisher’s combined probability test and effective agreement counts, and presents them through complementary visualisations that reveal where tools agree, where they diverge, and which findings depend on a single tool’s behaviour. The modular architecture is designed so that additional tools and enrichment paradigms can be integrated without changes to the comparison layer. Examiners Prof. Dimitrios Stravopodis, Prof. Ourania Tsitsilonis, Dr. Georgios Pavlopoulos link https://meet.google.com/gbg-mvby-dfn
MSc Thesis presentation of Mr. Sotirios Dimitriadis, Wednesday, February 25, 2026
On Wednesday 25 February 2026, at 12 pm, Mr Sotirios Dimitriadis, student of the postgraduate program “Data Science and Information Technologies”, will present his MSc thesis titled: “Predicting protein-membrane interfaces of peripheral membrane proteins using machine learning”
Title: “Predicting protein-membrane interfaces of peripheral membrane proteins using machine learning”
Sotirios Dimitriadis, DSIT MSc
Α.Μ.: 7115152300006
Location
Seminar Room ΙΣ2, Biomedical Research Foundation Academy of Athens, 4, Soranou Ephessiou, 115 27 Athens
Abstract
Peripheral membrane proteins are essential components in various biological activities, including cell differentiation, proliferation, and intercellular communication. Peripheral membrane proteins exhibit transient association with the lipid bilayer and are regulated with diverse mechanisms depending on their specific functions. Precise binding is required for cellular homeostasis, as protein-membrane attachment is responsible for the development of many disease pathologies. The study of these pathologies is often impeded because of the inherent difficulty of characterizing the protein-membrane interface through experimental techniques. For this reason, the specific membrane-binding domains of many peripheral membrane proteins remain unknown. This limitation has created a demand for the development of robust computational approaches for the prediction of protein-membrane interfaces. Current computational approaches, however, are often restricted by limited accuracy or excessive processing time. These constraints, when coupled with the scarcity of high-fidelity structural data, increase the development costs of small molecules intended to target the protein-membrane interface. To identify the protein-membrane interfaces we trained an ensemble machine learning model that predicts protein-membrane amino acids. The training of this model relied upon the consolidation of two distinct datasets. We curated the first dataset, compiling and manually updating high-quality experimental amino acid annotations through a comprehensive review of the recent literature. This information was supplemented by a second published dataset that incorporates interfacial binding site data from a wide variety of 9 protein superfamilies, which provides a broader representation of membrane-binding interactions. For the proteins in the consolidated dataset, we extracted physicochemical amino acid descriptors, geometrical features, and deep learning-based protein language model embeddings to capture diverse sequence and structural signals for the training of the model. To address the class imbalance and the lack of experimental annotation, we also augmented the protein-membrane interaction regions using a simple expansion strategy. Then, supervised Machine Learning training using binary classification ensued to predict the membrane-protein interacting amino acids. Model evaluation revealed that the highest predictive performance was achieved in a model trained on non-expanded experimental amino acid annotations using physicochemical and geometric features, yielding a Matthews Correlation Coefficient of 0.558 and an F1 Macro score of 0.776. The second-best model was trained using the consolidated dataset with expanded annotations, alongside a set of all available features. This work combines experimentally-curated annotations with expanded interface regions and evaluates the trade-off between physicochemical descriptors and protein language model embeddings.
Examiners
Dr. Zoe Cournia
Dr. Theodore Dalamagas
Dr. Konstantinos Vougas
Spring Semester scedule. February 16, 2026
Spring Semester starts on Monday, February 16, 2026
TIME-TABLE-SPRING-SEMESTER-DSIT-2025-26
AI in Medicine: From Models to Practice – Seminar- Biomedical Research Foundation Academy of Athens, January 26-28
This program combines lectures, hands-on workshops, and project work to give participants
both inspiration and practical skills. It is designed not only to showcase what is possible with AI
today, but also to spark participants’ curiosity and encourage them to pursue further learning
opportunities – ensuring that the Greek biomedical community continues to build expertise in
responsibly developing and applying AI in healthcare, bringing Greece to the forefront of current
innovation.
MIT-BRFAA AI in Medicine Program
