arXiv subjects

Sepinoud Azimi

Publications and source records attributed to Sepinoud Azimi.

11 recordsLinked to original sources

arXiv2026

Who Designs the Designer? Behavioural Architecture for GenAI in Education

AI in education is stuck between two failed responses: banning AI and building content-only tutors. Both fail because they ignore what decades of research has established: that personality, motivation, and emotional state shape learning outcomes as strongly as cognitive ability. This paper proposes behavioural architecture as an alternative. In the proposed architecture, the system adapts to how a student learns, not only to what they learn next. The student co-authors the record the system keeps, can read it, revise it, and revoke it. The designer role, what the system treats as true about the student, shifts from the AI vendor alone to a distribution among educator, student, and system. The paper argues that this architecture requires governance at EU level: the institution operating the system is the same one assessing the student, and individual institutions cannot provide the structural protections this configuration demands. Five empirical questions are proposed to test whether the architecture delivers on its claims. The contribution is naming a vacancy: the designer role in AI-in-education is currently unoccupied, and occupying it requires infrastructure that does not yet exist.

cs.CY↗

arXiv2025

Enabling Evolutionary Therapy in Metastatic Cancer Lacking Serum Biomarkers

Evolutionary therapy (ET) aims to steer tumor evolution by adjusting treatment timing and dosing to control rather than eradicate tumor burden. Clinical use requires reliable monitoring of tumor dynamics to inform mathematical models that guide therapy. In cancers such as metastatic castrate-resistant prostate cancer and relapsed platinum-sensitive ovarian cancer, ET models are informed by serial serum biomarkers. For cancers lacking reliable biomarkers, such as metastatic non-small cell lung cancer (NSCLC), radiographic imaging remains the primary method for treatment response assessment, typically using RECIST 1.1 criteria. RECIST, which tracks a few lesions with one-dimensional (1D) measurements and defines progression relative to the nadir, the smallest tumor burden recorded after treatment, was not designed to support ET. It may miss early regrowth, underrepresent tumor burden, and obscure disease trends. Using a virtual NSCLC patient model, we demonstrate that lesion selection and measurement dimensionality strongly affect progression detection. Two-dimensional metrics provide modest improvement, but only 3D volumetric measurements accurately capture both tumor burden and its dynamics, which are key requirements for ET. To support ET in cancers lacking biomarkers, response assessment must evolve beyond RECIST by integrating volumetric imaging, automated segmentation, and potentially liquid biopsies, alongside redefining progression criteria to enable adaptive, patient-centered treatments.

q-bio.PE↗

arXiv2025

Teaching the social media generation: rethinking learning without sacrificing quality

The rise of social media and AI tools has reshaped how students engage with learning, process information, and build trust in educational content. This generation prefers short, visual materials and fast feedback but often struggles with focus, critical thinking, and deep learning. Educators face the challenge of adapting teaching methods to these habits without lowering academic standards. This study presents a blended learning redesign of a first-year technical course at a Dutch university. Key features included short whiteboard videos before class, hands-on teamwork during class, narrative-style handouts to reinforce learning, in-class draft assignments without AI, and weekly anonymous feedback to adjust in real time. The results were promising: attendance increased by nearly 50%, and none of the regularly attending students failed the exam. Students found the videos useful but emphasized that in-person sessions were essential for understanding the material. While some resisted the shift in expectations, most appreciated the structure, clarity, and opportunities for active learning. This case suggests that combining digital familiarity with clear expectations and active support can help meet students where they are, while still challenging them to grow.

cs.CY↗

arXiv2025

CALF: A Conditionally Adaptive Loss Function to Mitigate Class-Imbalanced Segmentation

Imbalanced datasets pose a considerable challenge in training deep learning (DL) models for medical diagnostics, particularly for segmentation tasks. Imbalance may be associated with annotation quality limited annotated datasets, rare cases, or small-scale regions of interest (ROIs). These conditions adversely affect model training and performance, leading to segmentation boundaries which deviate from the true ROIs. Traditional loss functions, such as Binary Cross Entropy, replicate annotation biases and limit model generalization. We propose a novel, statistically driven, conditionally adaptive loss function (CALF) tailored to accommodate the conditions of imbalanced datasets in DL training. It employs a data-driven methodology by estimating imbalance severity using statistical methods of skewness and kurtosis, then applies an appropriate transformation to balance the training dataset while preserving data heterogeneity. This transformative approach integrates a multifaceted process, encompassing preprocessing, dataset filtering, and dynamic loss selection to achieve optimal outcomes. We benchmark our method against conventional loss functions using qualitative and quantitative evaluations. Experiments using large-scale open-source datasets (i.e., UPENN-GBM, UCSF, LGG, and BraTS) validate our approach, demonstrating substantial segmentation improvements. Code availability: https://anonymous.4open.science/r/MICCAI-Submission-43F9/.

eess.IV↗

arXiv2025

Kolmogorov-Arnold Networks and Evolutionary Game Theory for More Personalized Cancer Treatment

Personalized cancer treatment is revolutionizing oncology by leveraging precision medicine and advanced computational techniques to tailor therapies to individual patients. Despite its transformative potential, challenges such as limited generalizability, interpretability, and reproducibility of predictive models hinder its integration into clinical practice. Current methodologies often rely on black-box machine learning models, which, while accurate, lack the transparency needed for clinician trust and real-world application. This paper proposes the development of an innovative framework that bridges Kolmogorov-Arnold Networks (KANs) and Evolutionary Game Theory (EGT) to address these limitations. Inspired by the Kolmogorov-Arnold representation theorem, KANs offer interpretable, edge-based neural architectures capable of modeling complex biological systems with unprecedented adaptability. Their integration into the EGT framework enables dynamic modeling of cancer progression and treatment responses. By combining KAN's computational precision with EGT's mechanistic insights, this hybrid approach promises to enhance predictive accuracy, scalability, and clinical usability.

cs.LG↗

arXiv2024

Know Their Name: An Anthology of Great Women in the Shadows

There has been a long history of women innovators producing outstanding contributions to society and public benefit yet having their work passed over or sidelined or attributed to male colleagues. This phenomenon has been coined the Matilda Effect. The amendments to the record of human achievements are now taking place, with an increasing pace in recent times due to greater social enlightenment and awareness and the interest in social justice. However, there remains a gap that must be addressed. In this article, we demonstrate the disparity in scientific recognition for a handful of case studies through search data collected via Google Trends and plotted as time-series figures and choropleth maps. Search trends reflect a noticeable divergence between recognition of female and male innovators. However, we note that in more well-known cases of the Matilda Effect, such as the historical account of Rosalind Franklin vs. James Watson and Francis Crick, the differences become less pronounced, emphasizing the importance of publicizing recognition. In response to this revelation, this article presents the stories of several women innovators and their great achievements. We identify the truth behind several discoveries and inventions, while revealing the full nature of this historical problem of social exclusion.

physics.soc-ph↗

arXiv2022

Identifying gender bias in blockbuster movies through the lens of machine learning

The problem of gender bias is highly prevalent and well known. In this paper, we have analysed the portrayal of gender roles in English movies, a medium that effectively influences society in shaping people's beliefs and opinions. First, we gathered scripts of films from different genres and derived sentiments and emotions using natural language processing techniques. Afterwards, we converted the scripts into embeddings, i.e. a way of representing text in the form of vectors. With a thorough investigation, we found specific patterns in male and female characters' personality traits in movies that align with societal stereotypes. Furthermore, we used mathematical and machine learning techniques and found some biases wherein men are shown to be more dominant and envious than women, whereas women have more joyful roles in movies. In our work, we introduce, to the best of our knowledge, a novel technique to convert dialogues into an array of emotions by combining it with Plutchik's wheel of emotions. Our study aims to encourage reflections on gender equality in the domain of film and facilitate other researchers in analysing movies automatically instead of using manual approaches.

cs.CL↗

arXiv2022

SimuShips -- A High Resolution Simulation Dataset for Ship Detection with Precise Annotations

Obstacle detection is a fundamental capability of an autonomous maritime surface vessel (AMSV). State-of-the-art obstacle detection algorithms are based on convolutional neural networks (CNNs). While CNNs provide higher detection accuracy and fast detection speed, they require enormous amounts of data for their training. In particular, the availability of domain-specific datasets is a challenge for obstacle detection. The difficulty in conducting onsite experiments limits the collection of maritime datasets. Owing to the logistic cost of conducting on-site operations, simulation tools provide a safe and cost-efficient alternative for data collection. In this work, we introduce SimuShips, a publicly available simulation-based dataset for maritime environments. Our dataset consists of 9471 high-resolution (1920x1080) images which include a wide range of obstacle types, atmospheric and illumination conditions along with occlusion, scale and visible proportion variations. We provide annotations in the form of bounding boxes. In addition, we conduct experiments with YOLOv5 to test the viability of simulation data. Our experiments indicate that the combination of real and simulated images improves the recall for all classes by 2.9%.

cs.CV↗

arXiv2021

Evolutionary computational platform for the automatic discovery of nanocarriers for cancer treatment

We present the EVONANO platform for the evolution of nanomedicines with application to anti-cancer treatments. EVONANO includes a simulator to grow tumours, extract representative scenarios, and then simulate nanoparticle transport through these scenarios to predict nanoparticle distribution. The nanoparticle designs are optimised using machine learning to efficiently find the most effective anti-cancer treatments. We demonstrate our platform with two examples optimising the properties of nanoparticles and treatment to selectively kill cancer cells over a range of tumour environments.

cs.NE↗

arXiv2021

On the Verification and Validation of AI Navigation Algorithms

This paper explores the state of the art on to methods to verify and validate navigation algorithms for autonomous surface ships. We perform a systematic mapping study to find research works published in the last 10 years proposing new algorithms for autonomous navigation and collision avoidance and we have extracted what verification and validation approaches have been applied on these algorithms. We observe that most research works use simulations to validate their algorithms. However, these simulations often involve just a few scenarios designed manually. This raises the question if the algorithms have been validated properly. To remedy this, we propose the use of a systematic scenario-based testing approach to validate navigation algorithms extensively.

cs.AI↗

arXiv2020

Improving Students Performance in Small-Scale Online Courses -- A Machine Learning-Based Intervention

The birth of massive open online courses (MOOCs) has had an undeniable effect on how teaching is being delivered. It seems that traditional in class teaching is becoming less popular with the young generation, the generation that wants to choose when, where and at what pace they are learning. As such, many universities are moving towards taking their courses, at least partially, online. However, online courses, although very appealing to the younger generation of learners, come at a cost. For example, the dropout rate of such courses is higher than that of more traditional ones, and the reduced in person interaction with the teachers results in less timely guidance and intervention from the educators. Machine learning (ML) based approaches have shown phenomenal successes in other domains. The existing stigma that applying ML based techniques requires a large amount of data seems to be a bottleneck when dealing with small scale courses with limited amounts of produced data. In this study, we show not only that the data collected from an online learning management system could be well utilized in order to predict students overall performance but also that it could be used to propose timely intervention strategies to boost the students performance level. The results of this study indicate that effective intervention strategies could be suggested as early as the middle of the course to change the course of students progress for the better. We also present an assistive pedagogical tool based on the outcome of this study, to assist in identifying challenging students and in suggesting early intervention strategies.

cs.CY↗