Nicola Altini retweeted
#scacchi È Nicola Altini il nuovo Campione italiano Rapid @federscacchi. Ha battuto Artem Gilevych e Simone De Filomeno. Ecco i nomi dei campioni per singola categoria: facebook.com/share/p/1EgV2bE…
Nicola Altini retweeted
In this work we assessed the link among Pathomics, Transcriptomics, and PDA mutations by using CLAM🧠. This multimodal and interpretable framework has been published on Computerized Medical Imaging and Graphics by @ElsevierConnect. Click on for details ⏩
sciencedirect.com/science/ar…

ALT Overview of Multimodal Processing Pipeline. Imaging data were analyzed using a CNN and two foundation models for feature extraction, followed by a CLAM model for classification. Two distinct dimensionality reduction techniques, namely DEG analysis and Deep AEs, were considered for transcriptomics. Finally, both transcriptomic and pathomic data were classified using ML models. The predictions from both branches were combined to produce a multimodal output, and attention-maps and SHAP values were visualized to interpret the logic behind model’s predictions.
Nicola Altini retweeted
Congratulations to Demis Hassabis with the first Nobel price. I remember Demis told me his goal is two Nobel prices and AI being able to answer ALL questions about our existance in his life time. Dont know if the second is achievable or even wishable, but first seems realistic 👍
Nicola Altini retweeted
Check out this recently published manuscript on TCIA datasets which combined radiomics, clinical and mutation data from CPTAC Pancreatic cancer subjects to study their prognoses!
p.s. Thank you @GianMaria_Zeit for following our citation guidelines perfectly!
Our last article about time-dependent explainable radiomic analysis from the multi-omic cohort CPTAC-PDA is available on @ElsevierConnect CMPB. Thanks to @TCIA_News for its great effort in sharing imaging data to let us do our hard but loved work.
sciencedirect.com/science/ar…
Nicola Altini retweeted
Our last article about time-dependent explainable radiomic analysis from the multi-omic cohort CPTAC-PDA is available on @ElsevierConnect CMPB. Thanks to @TCIA_News for its great effort in sharing imaging data to let us do our hard but loved work.
sciencedirect.com/science/ar…
Nicola Altini retweeted
🎊Welcome to meet us at our 1st International Online Conference on Bioengineering 🔗iocbe2024.sciforum.net/
📅Time: October 16-18
⏳ Abstract Submission Deadline: 17th July; via sciforum.net/user/submission…
#biomaterials #biomechanics #regenerativemedicine #tissueengineering #Nano
Nicola Altini retweeted
🔬 Excited to share the publication on "NDG-CAM: Nuclei Detection in Histopathology Images with Semantic Segmentation Networks and Grad-CAM"👉 mdpi.com/2306-5354/9/9/475
#nuclei_segmentation #histopathology #deep_learning #semantic_segmentation #instance_segmentation #detection
Nicola Altini retweeted
MLPs are so foundational, but are there alternatives? MLPs place activation functions on neurons, but can we instead place (learnable) activation functions on weights? Yes, we KAN! We propose Kolmogorov-Arnold Networks (KAN), which are more accurate and interpretable than MLPs.🧵
Nicola Altini retweeted
Happy to share out last deep learning-based tool for prognosis in DLBCL 🧠💻 -> SurvIAE: Survival prediction with Interpretable Autoencoders from Gene Expression Data 🧬, with @N_Altini, @SabCiavarella, @MariellaVeglia1 sciencedirect.com/science/ar…
Nicola Altini retweeted
Our poster presented by @N_Altini at #bits2023, Bioinformatics Italian Society, on XAI applied to DLBCL for predicting Early Progressors combining DEGs and clinical determinants. @SabCiavarella @MariellaVeglia1
Nicola Altini retweeted
Out now our letter 📩 published by @Hemasphere_EHA 🩸: “A Decision-tree Approach to Stratify DLBCL Risk Based on Stromal and Immune Microenvironment Determinants”
journals.lww.com/hemasphere/…
A great squad effort by @SabCiavarella @MariellaVeglia1 and colleagues from @PolibaOfficial
Nicola Altini retweeted
It’s a pleasure to share our paper on H&E normalization methods. A great result for a multidisciplinary team.
sciencedirect.com/science/ar…














