publications
Journal articles, conference papers, preprints and datasets, newest first. Links lead to the published version.
2026
- From visual saliency to anatomical explanation: generating textual diagnostic reports from brain MRIs with saliency-grounded LLMsPaul Valery Nguezet, Elie Tagne Fute, Yusuf Brima, and 2 more authorsBioData Mining, Sep 2026
The opaque nature of deep learning models remains a significant barrier to their clinical adoption in medical imaging. This paper presents a systematic empirical benchmark of visual saliency attribution, atlas-based visualisation, and LLM-based reporting for brain tumour MRI classification, integrated into a unified pipeline, leveraging large language models (LLMs) to deliver human-interpretable diagnostic narratives. The proposed framework operates through three coupled stages. First, nine CNN architectures are extended with a dual-output hybrid formulation that simultaneously optimises a classification head and a segmentation head, enabling spatially richer feature learning. Second, visual saliency attribution methods, namely Grad-CAM, Grad-CAM++, and ScoreCAM, are applied to generate class-discriminative heatmaps, which are subsequently refined into coarse binary masks via an adaptive percentile thresholding pipeline. Third, the resulting masks are projected onto the Harvard–Oxford cortical atlas as an illustrative anatomical overlay, offering an approximate visual reference for the tumour’s cortical neighbourhood, and the extracted findings are encoded into a structured JSON file that conditions three LLMs (Grok3, Mistral, and LLaMA) to generate coherent, radiological-style diagnostic reports. Evaluated on a dataset of 4,834 contrast-enhanced T1-weighted brain MRI images spanning three tumour classes, InceptionResNetV2 achieved the highest classification performance, and Grad-CAM++ yielded the best segmentation overlap. Among the language models, Grok3 led in lexical diversity and coherence, while LLaMA achieved the highest readability score. By integrating visual, anatomical, and linguistic modalities into a unified pipeline, the framework attempts to produce technically grounded and meaningfully interpretable explanations, offering a step toward more transparent artificial intelligence-assisted brain tumour diagnosis.
@article{nguezet2026saliencyllm, title = {From visual saliency to anatomical explanation: generating textual diagnostic reports from brain {MRIs} with saliency-grounded {LLMs}}, author = {Nguezet, Paul Valery and Fute, Elie Tagne and Brima, Yusuf and Azanguezet, Benoit Martin and Atemkeng, Marcellin}, journal = {BioData Mining}, year = {2026}, month = sep, publisher = {Springer Science and Business Media LLC}, issn = {1756-0381}, doi = {10.1186/s13040-026-00601-w}, } - Transformers for 3D medical image analysis: a systematic review of architectural innovations, performance, and clinical applicationsWandile Nhlapho, Marcellin Atemkeng, Yusuf Brima, and 4 more authorsArtificial Intelligence Review, Jul 2026
A systematic review of 138 studies examining Transformer-based architectures for 3D medical image analysis across MRI, CT, PET, and ultrasound. The review synthesizes architectural innovations, benchmark performance, clinical applications, and reproducibility challenges while highlighting emerging trends such as hybrid CNN–Transformer models, State Space Models, and diffusion-based Transformers.
@article{nhlapho2026transformers, title = {Transformers for 3D medical image analysis: a systematic review of architectural innovations, performance, and clinical applications}, author = {Nhlapho, Wandile and Atemkeng, Marcellin and Brima, Yusuf and Ravelo, Blaise and Bouwmans, Thierry and Lalléchère, Sébastien and Ndogmo, Jean-Claude}, journal = {Artificial Intelligence Review}, volume = {59}, number = {11}, year = {2026}, month = jul, publisher = {Springer Science and Business Media LLC}, issn = {1573-7462}, doi = {10.1007/s10462-026-11600-y}, } - An empirical study of machine learning robustness and scalability for imbalanced tabular clinical data in emergency and critical careYusuf Brima and Marcellin AtemkengScientific Reports, Jun 2026
Machine learning can support high-stakes decision-making in emergency and intensive care settings, but severe class imbalance in clinical data limits model reliability and biases predictions toward majority outcomes. We evaluate six model families, including classical methods (Decision Tree, Random Forest, XGBoost), deep learning approaches (TabNet), and tabular foundation models (TabICL, TabPFN v2.6), on MIMIC-IV-ED and eICU datasets across multiple clinical prediction tasks. Models are assessed using Macro F1-score, robustness to increasing imbalance, and computational efficiency. Results show dataset-dependent performance: TabPFN and TabICL perform strongly on MIMIC-IV-ED, while XGBoost leads on eICU. No single model dominates across all settings, but foundation models provide a favorable efficiency–performance trade-off and are increasingly competitive in imbalanced clinical prediction scenarios.
@article{brima2026robustness, title = {An empirical study of machine learning robustness and scalability for imbalanced tabular clinical data in emergency and critical care}, author = {Brima, Yusuf and Atemkeng, Marcellin}, journal = {Scientific Reports}, volume = {16}, number = {1}, year = {2026}, month = jun, publisher = {Springer Science and Business Media LLC}, issn = {2045-2322}, doi = {10.1038/s41598-026-56413-9}, } - A Multimodal Explainable Deep Learning Framework for Alzheimer’s Disease Diagnosis using 3D Magnetic Resonance Imaging and Clinical DataYusuf Brima, Marcellin Atemkeng, Lakshmana Rao Namamula, and 1 more author2026Under review at Cognitive Computation
@unpublished{brima2026multimodal, title = {A Multimodal Explainable Deep Learning Framework for Alzheimer's Disease Diagnosis using 3D Magnetic Resonance Imaging and Clinical Data}, author = {Brima, Yusuf and Atemkeng, Marcellin and Namamula, Lakshmana Rao and Vacavant, Antoine}, year = {2026}, archiveprefix = {arXiv}, note = {Under review at Cognitive Computation}, } - Few-shot Cross-country Generalization of Tabular Machine Learning and Foundation Models for Childhood Anemia Prediction under Distribution ShiftYusuf Brima, Marcellin Atemkeng, Lansana Hassim Kallon, and 4 more authors2026Under review at BioData Mining
Childhood anemia affects 40% of children aged 6–59 months globally and is driven by heterogeneous, context-dependent factors that challenge model generalization. We evaluate a transformer-based tabular foundation model (TabPFN v2.6) against classical methods (Logistic Regression, XGBoost, LightGBM) using DHS data from 16 countries (n=68,856) under cross-country and data-scarce settings. Across LOCO, reverse-LOCO, and few-shot evaluations, TabPFN shows superior performance in low-data regimes, with improved calibration (Brier: 0.042; ECE: 0.203) and strong discrimination. While full-data performance differences are modest, results indicate that population heterogeneity dominates predictive performance, and foundation models offer advantages for robust, low-resource global health prediction.
@unpublished{brima2026anemia, title = {Few-shot Cross-country Generalization of Tabular Machine Learning and Foundation Models for Childhood Anemia Prediction under Distribution Shift}, author = {Brima, Yusuf and Atemkeng, Marcellin and Kallon, Lansana Hassim and Niyukuri, David and Vacavant, Antoine and Saidu, Samuel and Chen, Ding-Geng}, year = {2026}, url = {https://arxiv.org/abs/2605.26589}, archiveprefix = {arXiv}, primaryclass = {cs.LG}, note = {Under review at BioData Mining}, } - Artificial intelligence in neglected tropical diseases: current applications, challenges, and opportunitiesYusuf Brima, Marcellin Atemkeng, Mikel K. Ngueajio, and 30 more authors2026Submitted to npj Digital Medicine
Neglected tropical diseases (NTDs) affect over one billion people globally and disproportionately impact low-resource tropical and subtropical regions. While artificial intelligence (AI) offers significant opportunities to strengthen NTD control, the scope, maturity, and translational readiness of existing applications remain unclear. We conducted a systematic review of AI methods applied to NTD-related problems using Scopus and Web of Science, following PRISMA guidelines. Across 289 studies, AI applications were primarily concentrated in diagnosis, surveillance, and prognosis, with most relying on supervised learning from imaging, clinical, or epidemiological data. However, external validation, calibration, and prospective evaluation were limited, and research was geographically concentrated in a small number of countries. Overall, despite rapid growth in the field, substantial gaps remain in methodological rigor, equity, and real-world deployment readiness, highlighting the need for more robust, transparent, and context-aware AI research in NTDs.
@unpublished{brima2026aintd, title = {Artificial intelligence in neglected tropical diseases: current applications, challenges, and opportunities}, author = {Brima, Yusuf and Atemkeng, Marcellin and Ngueajio, Mikel K. and Ngueabou, Yolande and Nguembang Fadja, Arnaud and Bonginkosi, Ndzamba and Bouwmans, Thierry and Chen, Ding-Geng and Nkouindja Tatchou, Daniel Nathan and Mwania, Elizabeth Mukusya and Fashoto, Stephen Gbenga and Fendji, Jean Louis Ebongue Kedieng and Jojo, Masixole and Kamara, Lukeman Sahid and Kallon, Lansana Hassim and Kargbo, Adam and Kitio, Linuce Demanou and Likassa, Habte Tadesse and Lalléchère, Sébastien and Makoni, Pedzisai Anotida and Mokhele, Shoeshoe and Mosima, Lerato Sarah and Odume, Oghenekaro Nelson and Ravelo, Blaise and Nnadozie, Chika Felicitas and Sandjo, Louis Pergaud and Saidu, Samuel and Sibiya, Ntethelelo and Soares, Bruno Gabriel Larsen and Tenkam, Herve Michel and Witika, Bwalya Angel and Zaza, Siphendulwe and Siwe Noundou, Xavier}, year = {2026}, note = {Submitted to npj Digital Medicine}, } - From Benchmark to Bedside: A Roadmap for Trustworthy Artificial Intelligence Across African Healthcare ContextsYusuf Brima, Marcellin Atemkeng, Nicole Oyetunji, and 2 more authors2026Proposed long article for the CACM Africa Special Section (Societal Impact: Healthcare)
Artificial intelligence (AI) is advancing across the healthcare continuum, from early diagnosis and risk stratification to public health surveillance. Across Africa, researchers are applying state-of-the-art AI models to local health data at all scales, from community clinics to tertiary referral hospitals. These efforts show both remarkable promise and a clear caution: strong benchmark scores often fail to transfer to populations and workflows the models were not trained on. This work asks how healthcare AI in Africa can move from promising, isolated pilots to safe, locally governed and sustainable use. We argue that the next step is to connect these use cases to a broader, locally governable ecosystem. African data realities are already testing global models: recent studies in fetal ultrasound across five countries, brain MRI in Nigeria, cervical CT in Zambia, diabetic retinopathy screening in Kenya and dermatology in Uganda reveal downsides of transfer, under-representation, and explanations not always faithful to model behaviour. At the same time, work in tuberculosis screening, ultraportable X-ray, cough-based screening and smartphone ultrasound shows how AI may support services facing severe workforce and infrastructure constraints, while reminding us that accuracy alone does not establish clinical utility. We bring these developments together in a five-stage roadmap for translating healthcare AI: development, evaluation, deployment, scaling and, crucially, sustainment. Across the cycle, we identify three foundations that must move together: responsibly governed, interoperable data aligned with the FAIR principles; African agency over data, models, compute infrastructure and research priorities, including through federated and emerging multimodal approaches; and trust built through locally relevant validation, transparent reporting, regulation and prospective real-world evidence. Fit-for-purpose technology, affordable workflows, interdisciplinary capacity and sustained funding are not afterthoughts but preconditions for lasting impact. This article synthesises examples across countries and programmes, including AFRICAI and AFRICAI-RI, and connects them to the African Union’s Data Policy Framework and Continental AI Strategy, WHO guidance, and consensus frameworks such as FUTURE-AI. Rather than treating Africa as one health system or proposing a single model for every setting, we make the case for locally governable AI systems evaluated in tandem with the communities and services expected to use and sustain them. Africa carries a large share of the global disease burden while facing acute shortages of health workers, diagnostic capacity and infrastructure. AI could help extend scarce expertise, but only if tools are validated on local populations, fit real clinical workflows and are governed by African institutions, rather than imported as finished products. Globally, models developed or validated across diverse African contexts can expose blind spots in supposedly general-purpose AI, while a stronger African research ecosystem can contribute to more robust, equitable and clinically useful healthcare AI. Africa’s genetic, epidemiological and health-system diversity makes it a demanding testbed for how well AI generalises, with lessons for other low- and middle-income settings and for high-income health systems serving diverse populations.
@unpublished{brima2026benchmark, title = {From Benchmark to Bedside: A Roadmap for Trustworthy Artificial Intelligence Across African Healthcare Contexts}, author = {Brima, Yusuf and Atemkeng, Marcellin and Oyetunji, Nicole and Nhlapho, Wandile and Fendji, Jean Louis Ebongue Kedieng}, year = {2026}, note = {Proposed long article for the CACM Africa Special Section (Societal Impact: Healthcare)}, }
2025
- An unsupervised XAI framework for dementia detection with context enrichmentDevesh Singh, Yusuf Brima, Fedor Levin, and 44 more authorsScientific Reports, Nov 2025
Explainable Artificial Intelligence (XAI) methods enhance the diagnostic efficiency of clinical decision support systems by making the predictions of a convolutional neural network’s (CNN) on brain imaging more transparent and trustworthy. However, their clinical adoption is limited due to limited validation of the explanation quality. Our study introduces a framework that evaluates XAI methods by integrating neuroanatomical morphological features with CNN-generated relevance maps for disease classification. We trained a CNN using brain MRI scans from six cohorts: ADNI, AIBL, DELCODE, DESCRIBE, EDSD, and NIFD (N = 3253), including participants that were cognitively normal, with amnestic mild cognitive impairment, dementia due to Alzheimer’s disease and frontotemporal dementia. Clustering analysis benchmarked different explanation space configurations by using morphological features as proxy-ground truth. We implemented three post-hoc explanations methods: (i) by simplifying model decisions, (ii) explanation-by-example, and (iii) textual explanations. A qualitative evaluation by clinicians (N = 6) was performed to assess their clinical validity. Clustering performance improved in morphology enriched explanation spaces, improving both homogeneity and completeness of the clusters. Post hoc explanations by model simplification largely delineated converters and stable participants, while explanation-by-example presented possible cognition trajectories. Textual explanations gave rule-based summarization of pathological findings. Clinicians’ qualitative evaluation highlighted challenges and opportunities of XAI for different clinical applications. Our study refines XAI explanation spaces and applies various approaches for generating explanations. Within the context of AI-based decision support system in dementia research we found the explanations methods to be promising towards enhancing diagnostic efficiency, backed up by the clinical assessments.
@article{singh2025xaidementia, title = {An unsupervised {XAI} framework for dementia detection with context enrichment}, author = {Singh, Devesh and Brima, Yusuf and Levin, Fedor and Becker, Martin and Hiller, Bjarne and Hermann, Andreas and Villar-Munoz, Irene and Beichert, Lukas and Bernhardt, Alexander and Buerger, Katharina and Butryn, Michaela and Dechent, Peter and Düzel, Emrah and Ewers, Michael and Fliessbach, Klaus and Freiesleben, Silka D. and Glanz, Wenzel and Hetzer, Stefan and Janowitz, Daniel and Görß, Doreen and Kilimann, Ingo and Kimmich, Okka and Laske, Christoph and Levin, Johannes and Lohse, Andrea and Luesebrink, Falk and Munk, Matthias and Perneczky, Robert and Peters, Oliver and Preis, Lukas and Priller, Josef and Prudlo, Johannes and Prychynenko, Diana and Rauchmann, Boris S. and Rostamzadeh, Ayda and Roy-Kluth, Nina and Scheffler, Klaus and Schneider, Anja and Droste zu Senden, Louise and Schott, Björn H. and Spottke, Annika and Synofzik, Matthis and Wiltfang, Jens and Jessen, Frank and Weber, Marc-André and Teipel, Stefan J. and Dyrba, Martin}, journal = {Scientific Reports}, volume = {15}, number = {1}, pages = {39554}, year = {2025}, month = nov, publisher = {Springer Science and Business Media LLC}, issn = {2045-2322}, doi = {10.1038/s41598-025-26227-2}, } - A systematic review of low-rank and local low-rank matrix approximation in big data medical imagingSisipho Hamlomo, Marcellin Atemkeng, Yusuf Brima, and 2 more authorsNeural Computing and Applications, Jun 2025
The large volume and complexity of medical imaging datasets pose significant challenges for storage, transmission, and processing. To address these issues, low-rank matrix approximation (LORMA) and its derivative, local LORMA (LLORMA), have shown promising potential. This paper presents a comprehensive literature review of the application of LORMA and LLORMA across various imaging modalities and examines the challenges and limitations of existing methods. Notably, since 2015, there has been a significant shift toward a preference for LLORMA in the medical imaging field, demonstrating its effectiveness in capturing complex structures in medical data compared to LORMA. Given the limitations of shallow similarity methods in LLORMA, we propose incorporating advanced semantic image segmentation to improve the accuracy of similarity measurement. We further explain how this approach can be utilized to identify similar patches and assess its feasibility in medical imaging applications. We observe that LORMA and LLORMA have primarily been applied to unstructured medical data, and we suggest extending their use to other types of medical data, including structured and semi-structured formats. This paper also explores how LORMA and LLORMA can be adapted for regular data with missing entries, considering the impact of inaccuracies in predicting these missing values and their consequences. In addition, we examine the effect of patch size and suggest using random search (RS) to identify the optimal patch size. To further enhance feasibility, we propose a hybrid approach combining Bayesian optimization and RS, which could improve the application of LORMA and LLORMA in medical imaging.
@article{hamlomo2025lowrank, title = {A systematic review of low-rank and local low-rank matrix approximation in big data medical imaging}, author = {Hamlomo, Sisipho and Atemkeng, Marcellin and Brima, Yusuf and Nunhokee, Chuneeta and Baxter, Jeremy}, journal = {Neural Computing and Applications}, volume = {37}, number = {16}, pages = {9481--9536}, year = {2025}, month = jun, publisher = {Springer Science and Business Media LLC}, issn = {1433-3058}, doi = {10.1007/s00521-025-11055-2}, } - Disentangled Representation Learning in Speech and VocalizationYusuf BrimaUniversität Osnabrück, 2025
@phdthesis{brima2025thesis, title = {Disentangled Representation Learning in Speech and Vocalization}, author = {Brima, Yusuf}, school = {Universität Osnabrück}, year = {2025}, }
2024
- Saliency-driven explainable deep learning in medical imaging: bridging visual explainability and statistical quantitative analysisYusuf Brima and Marcellin AtemkengBioData Mining, Jun 2024
Deep learning shows great promise for medical image analysis but often lacks explainability, hindering its adoption in healthcare. Attribution techniques that explain model reasoning can potentially increase trust in deep learning among clinical stakeholders. In the literature, much of the research on attribution in medical imaging focuses on visual inspection rather than statistical quantitative analysis.
@article{brima2024saliency, title = {Saliency-driven explainable deep learning in medical imaging: bridging visual explainability and statistical quantitative analysis}, author = {Brima, Yusuf and Atemkeng, Marcellin}, journal = {BioData Mining}, volume = {17}, number = {1}, pages = {18}, year = {2024}, month = jun, publisher = {Springer Science and Business Media LLC}, issn = {1756-0381}, doi = {10.1186/s13040-024-00370-4}, } - Bridging the Gap: Exploring Interpretability in Deep Learning Models for Brain Tumor Detection and Diagnosis from MRI ImagesWandile Nhlapho, Marcellin Atemkeng, Yusuf Brima, and 1 more authorInformation, Mar 2024
The advent of deep learning (DL) has revolutionized medical imaging, offering unprecedented avenues for accurate disease classification and diagnosis. DL models have shown remarkable promise for classifying brain tumors from Magnetic Resonance Imaging (MRI) scans. However, despite their impressive performance, the opaque nature of DL models poses challenges in understanding their decision-making mechanisms, particularly crucial in medical contexts where interpretability is essential. This paper explores the intersection of medical image analysis and DL interpretability, aiming to elucidate the decision-making rationale of DL models in brain tumor classification. Leveraging ten state-of-the-art DL frameworks with transfer learning, we conducted a comprehensive evaluation encompassing both classification accuracy and interpretability. These models underwent thorough training, testing, and fine-tuning, resulting in EfficientNetB0, DenseNet121, and Xception outperforming the other models. These top-performing models were examined using adaptive path-based techniques to understand the underlying decision-making mechanisms. Grad-CAM and Grad-CAM++ highlighted critical image regions where the models identified patterns and features associated with each class of the brain tumor. The regions where the models identified patterns and features correspond visually to the regions where the tumors are located in the images. This result shows that DL models learn important features and patterns in the regions where tumors are located for decision-making.
@article{info15040182, title = {Bridging the Gap: Exploring Interpretability in Deep Learning Models for Brain Tumor Detection and Diagnosis from {MRI} Images}, author = {Nhlapho, Wandile and Atemkeng, Marcellin and Brima, Yusuf and Ndogmo, Jean-Claude}, journal = {Information}, volume = {15}, number = {4}, pages = {182}, year = {2024}, month = mar, publisher = {MDPI AG}, issn = {2078-2489}, doi = {10.3390/info15040182}, url = {https://www.mdpi.com/2078-2489/15/4/182}, } - Understanding Self-Supervised Learning of Speech Representation via Invariance and Redundancy ReductionYusuf Brima, Ulf Krumnack, Simone Pika, and 1 more authorInformation, Feb 2024
This paper explores self-supervised learning for speech representation, leveraging redundancy reduction techniques to learn robust representations. It adapts the Barlow Twins framework to speech processing and evaluates its effectiveness in various downstream tasks.
@article{brima2024sslspeech, title = {Understanding Self-Supervised Learning of Speech Representation via Invariance and Redundancy Reduction}, author = {Brima, Yusuf and Krumnack, Ulf and Pika, Simone and Heidemann, Gunther}, journal = {Information}, volume = {15}, number = {2}, pages = {114}, year = {2024}, month = feb, publisher = {MDPI AG}, issn = {2078-2489}, doi = {10.3390/info15020114}, } - Learning Disentangled Audio Representations through Controlled SynthesisYusuf Brima, Ulf Krumnack, Simone Pika, and 1 more authorIn ICLR 2024 Tiny Papers Track, 2024
This paper introduces a novel approach to learning disentangled speech representations through controlled synthesis, enabling better interpretability and generalization in speech processing tasks.
@inproceedings{brima2024disentangledaudio, title = {Learning Disentangled Audio Representations through Controlled Synthesis}, author = {Brima, Yusuf and Krumnack, Ulf and Pika, Simone and Heidemann, Gunther}, booktitle = {ICLR 2024 Tiny Papers Track}, year = {2024}, url = {https://openreview.net/forum?id=Fn9ORH8PLl}, }
2023
- A Mathematical Framework for Understanding Recognition SystemsYusuf BrimabioRxiv, Jun 2023
We propose a theoretical framework for biological recognition, integrating category theory, information theory, dynamical systems modeling, and optimization theory. This framework defines individual recognition (IR) and class-level recognition (CR) as functors between stimuli and responses. We identify five conditions for robust IR systems, termed "signature systems," and model them as attractor states, offering insights into communication and language evolution.
@article{brima2023mir, title = {A Mathematical Framework for Understanding Recognition Systems}, author = {Brima, Yusuf}, journal = {bioRxiv}, year = {2023}, month = jun, publisher = {openRxiv}, doi = {10.1101/2023.06.08.544240}, } - Self-Supervised Learning of Speech Representation via Redundancy ReductionYusuf BrimaIn KI 2023 Doctoral Consortium, 2023
This work aims to investigate a novel self-supervised learning (SSL) method for speech representation that leverages redundancy reduction techniques to learn robust representations capturing speaker characteristics. Our proposed approach builds upon the Barlow Twins framework, introduced in computer vision, and we adapt it to speech processing. The primary objective is to assess the quality of the learned representations through comprehensive evaluations of various downstream tasks, including speaker identification, gender recognition, emotion recognition, and more.
@inproceedings{brima2023sslrrd, title = {Self-Supervised Learning of Speech Representation via Redundancy Reduction}, author = {Brima, Yusuf}, booktitle = {KI 2023 Doctoral Consortium}, year = {2023}, publisher = {Gesellschaft für Informatik e.V.}, doi = {10.18420/ki2023-dc-02}, url = {https://dl.gi.de/handle/20.500.12116/42402}, } - Learning Disentangled Speech RepresentationsYusuf Brima, Ulf Krumnack, Simone Pika, and 1 more authorIn New in Machine Learning Workshop, NeurIPS 2023, 2023
This study investigates methods for learning disentangled representations of speech signals, focusing on separating different speech-related factors such as speaker identity, linguistic content, and emotional state. The work applies novel deep learning approaches to improve representation learning in speech processing.
@inproceedings{brima2023disentangledspeech, title = {Learning Disentangled Speech Representations}, author = {Brima, Yusuf and Krumnack, Ulf and Pika, Simone and Heidemann, Gunther}, booktitle = {New in Machine Learning Workshop, NeurIPS 2023}, year = {2023}, url = {https://openreview.net/forum?id=3ox1TfKeRF}, }
2022
- Deep Transfer Learning for Brain Magnetic Resonance Image Multi-class ClassificationYusuf Brima, Mosaddek Hossain Kamal Tushar, Upama Kabir, and 1 more authorDhaka University Journal of Applied Science and Engineering, Jun 2022
This paper presents a deep transfer learning approach for multi-class classification of brain MRI images, evaluating its effectiveness in differentiating various neurological conditions.
@article{brima2022transferbrainmri, title = {Deep Transfer Learning for Brain Magnetic Resonance Image Multi-class Classification}, author = {Brima, Yusuf and Tushar, Mosaddek Hossain Kamal and Kabir, Upama and Islam, Tariqul}, journal = {Dhaka University Journal of Applied Science and Engineering}, volume = {6}, number = {2}, pages = {14--29}, year = {2022}, month = jun, publisher = {Bangladesh Journals Online (JOL)}, issn = {2218-7413}, doi = {10.3329/dujase.v6i2.59215}, }
2021
- Transfer Learning for the Detection and Diagnosis of Types of Pneumonia including Pneumonia Induced by COVID-19 from Chest X-ray ImagesYusuf Brima, Marcellin Atemkeng, Stive Tankio Djiokap, and 2 more authorsDiagnostics, Aug 2021
This study explores the use of transfer learning for detecting and diagnosing different types of pneumonia, including COVID-19-induced pneumonia, from chest X-ray images. The work evaluates the performance of various deep learning models in assisting medical professionals with automated diagnostics.
@article{brima2021pneumoniadetection, title = {Transfer Learning for the Detection and Diagnosis of Types of Pneumonia including Pneumonia Induced by {COVID-19} from Chest X-ray Images}, author = {Brima, Yusuf and Atemkeng, Marcellin and Tankio Djiokap, Stive and Ebiele, Jaures and Tchakounté, Franklin}, journal = {Diagnostics}, volume = {11}, number = {8}, pages = {1480}, year = {2021}, month = aug, publisher = {MDPI AG}, issn = {2075-4418}, doi = {10.3390/diagnostics11081480}, } - Brain MRI DatasetYusuf Brima, Mosaddek Hossain Kamal Tushar, Upama Kabir, and 1 more author2021
Published a dataset that was curated in collaboration between the Computer Science and Engineering Department, University of Dhaka and the National Institute of Neuroscience, Bangladesh. It comprise 5,285 T1-weighted contrast- enhanced brain MRI images belonging to 38 categories.
@misc{brima2021mridata, title = {Brain {MRI} Dataset}, author = {Brima, Yusuf and Tushar, Mosaddek Hossain Kamal and Kabir, Upama and Islam, Tariqul}, year = {2021}, publisher = {figshare}, doi = {10.6084/m9.figshare.14778750.v2}, url = {https://figshare.com/articles/dataset/Brain_MRI_Dataset/14778750/2}, }