Open-access, independent, minimum fee journal. Co-founded by @arbtal,@ja_schnabel,@mertrory,@wmwells3,@MarcNiethammer,@adriandalca
Joined January 2020
- Tweets514
- Following380
- Followers3.6K
- Likes273
Pinned Tweet
We have a new low-frequency email list. Please consider signing up if you want to hear about our important announcements.
melba-journal.org/contact.ht…
🚨 New publication alert:
📢 ”A multi-site loiasis dataset of whole blood videos.”
🖊️ L Djune-Yemeli, R Zoleko-Manego, C B Delahunt, D F Fankam, Y Y Nzeuhang, ..., E Spencer, D Banik, M D Keller, K N Fisher, D A Fletcher, I I Bogoch, A-L Le Ny, S Deda David, J Kamgno.
⬇️
🎯 Authors introduce the first large-scale dataset for AI-based detection of Loa loa parasites in blood, featuring over 33,000 microscopy videos from nearly 2,000 patients in Cameroon and Gabon.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”RCC-AID: Renal Cell Carcinoma AI Dataset for Medical Imaging Research.”
🖊️ S de Boer, H Häntze, S Ziegelmayer, T Russo, B van Ginneken, M Prokop, K Bressem, A Hering.
⬇️
🎯 Authors introduce a public dataset of 129 CT scans from 91 patients with renal cell carcinoma, featuring voxel-level annotations of kidneys, tumors, and cysts, supporting segmentation, detection, and classification tasks.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”CFB-GBM v2.0: An Augmented Longitudinal Dataset for Multi-Modal Glioblastoma Segmentation, Radiomics, and RANO Progression Tracking.”
🖊️ A G Leclerq, N N Moreau, H Audebert, A Nassar, T Cochin, ..., J Lacroix, R Rouzier, A Corroyer-Dulmont
⬇️
🎯 Authors introduce a longitudinal multimodal MRI dataset of 264 glioblastoma patients, with tumor delineations completed across 97% of available timepoints, as well as treatment-response labels, brain masks, and radiomic features.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”A Histopathological Image Dataset for Oral Squamous Cell Carcinoma and Normal Tissue Analysis.”
🖊️ S Bahaa.
⬇️
🎯 Authors introduce an expert-labeled histopathology dataset for oral squamous cell carcinoma detection, with 1,714 images from 115 patients and paired low- and high-magnification images supporting multi-scale analysis.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”CT-Bench: A Comprehensive Benchmark for Multimodal AI in Computed Tomography Analysis.”
🖊️ Q Zhu, Q Jin, T S Mathai, Y Fang, …, P T S Balamuralikrishna, K C Wang, R M Summers, Z Lu.
⬇️
🎯 Authors introduce CT-Bench, a multimodal CT lesion resource and visual question answering benchmark with over 20,000 annotated lesions, supporting image–text alignment, localization, attributes, and lesion-level reasoning tasks.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”ABFR-KAN: Kolmogorov-Arnold Networks for Functional Brain Analysis.”
🖊️ T Ward, A Imran.
⬇️
🎯 Authors propose an approach combining anatomically adaptive brain representations with Kolmogorov–Arnold Networks to reduce atlas-related bias in functional connectivity analysis, improving downstream classification.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”Geodesic-informed Generative Diffusion Model For Topology-preserved Image Video Generation.”
🖊️ N Wu, N Jayakumar, J Xing, M Zhang.
⬇️
🎯 Authors introduce a diffusion framework that generates images through topology-preserving geodesic deformations, explicitly modeling shape changes. The approach produces realistic images while preserving anatomical structure.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”Benchmarking Self-Supervised Learning Methods for Accelerated MRI Reconstruction.”
🖊️ A Wang, S McDonagh, M Davies.
⬇️
🎯 Authors introduce SSIBench, an open-source framework benchmarking 21 self-supervised MRI reconstruction methods across seven realistic scenarios without fully sampled ground truth.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”The MYOSAIQ Challenge: Myocardial Segmentation with Automated Infarct Quantification.”
🖊️ O Bernard, W A Romero R, C Bouton, C Goujat, H Jung Ling, P-M Jodoin, ..., P Clarysse, L Belle, T Bochaton, N Mewton, M Viallon, P Croisille.
⬇️
🎯 Authors present the MYOSAIQ challenge, benchmarking myocardial infarct segmentation across a large multicenter cardiac MRI dataset. Results show that well-designed U-Net methods outperform fully automated foundation models.
🔎 Free to read: doi.org/10.59275/j.melba.202…
🚨 New publication alert:
📢 ”Patient Pose Assessment Using a CT-Based Framework for Synthetic Data Generation.”
🖊️ M Laufer, D Mairhöfer, M Sieren, H Gerdes, F Leal dos Reis, A Bischof, T Käster, E Barth, J Barkhausen, T Martinetz.
⬇️
🎯 Authors propose a framework to generate synthetic depth images from CT scans for AI-based patient positioning in radiography. Pretraining on synthetic data improves pose assessment on real radiographs by up to 11 percentage points.
🔎 Free to read: doi.org/10.59275/j.melba.202…
