PURR.AI’s Post

I'm happy to share the expanded abstract titled "Convolutional Neural Network for Clinical Diagnosis Support of Spondylolisthesis Through Imaging: Expanded Abstract," part of my Master's project at Universidade Federal de Ciências da Saúde de Porto Alegre, has been accepted at the IC-AI Health event: Responsible Artificial Intelligence in Education, Research, and Practices in Healthcare. *** Estou feliz em compartilhar que o resumo expandido intitulado "Convolutional Neural Network for Clinical Diagnosis Support of Spondylolisthesis Trhough Imaging: Expanded Abstract", parte do meu projeto de Mestrado na UFCSPA, foi aprovado no evento CI-IA Saúde: Inteligência Artificial responsável no ensino, pesquisa e práticas em saúde. #CIIASaúde #UFCSPA #ArtificialIntelligence

📢 A new year (2025) gift is available for you I feel very happy to say that: 🌟 The fourth journal article from my MS + PhD thesis has been published with collaborator Prof. Amalesh Jana from Montana State University-Bozeman, USA, and my MS + PhD supervisor, Prof. G. R. Dodagoudar, Indian Institute of Technology, Madras, India, in one of the prestigious journals, International Journal for Numerical and Analytical Methods in Geomechanics, Wiley, in this field. The authors would like to thank the editors and two anonymous reviewers for their useful suggestions, and considered this article a useful and interesting. Special thanks to them for the timely review.🌟 🚀 Key highlights: In this study, the data-driven constitutive models are developed using only laboratory test databases and deep learning techniques. The laboratory database was prepared by conducting CDSS tests on reconstituted sand, i.e., PDX sand. The stacked long short-term memory (LSTM) network and its variants are considered for developing the predictive models of the shear strain and excess pore pressure ratio time histories. In summary, it is observed that though the trained models predicted the time histories reasonably well; however, they struggled to predict the hysteresis loops at higher cycles. Therefore, more research is needed to verify and enhance the predictability of existing AI-based models in the future before using them in practice for simulating cyclic response. 🌟 Cite this article: Jas, K., Jana, A., Dodagoudar, G. R. (2025). Evaluation and future prospects of data-driven intelligence-based framework for predicting cyclic behavior of reconstituted sand. International Journal for Numerical and Analytical Methods in Geomechanics, 0:1-25. https://lnkd.in/dzSAxsvV. I have attached the DOI of this article. Kindly download and read it. If you find it interesting and useful research, share it with your friends and students. Journal information: Impact factor: 3.4 (Q1); Cite score: 6.4; Acceptance rate: 34% You can recommend in Researchgate: https://lnkd.in/dJaVTMXn. 🌟My scholarly profiles: ResearchGate: https://lnkd.in/g9CkMSca.   ORCID: https://lnkd.in/ekCkgQZg. Website: https://lnkd.in/dapnfFh4. Twitter: https://lnkd.in/dN6H6pRT. #JustPublished #Wiley Wiley In Research #IJNAMG #PaperPromotion #research #phd #engineering #ArtificialIntelligence #machinelearning #AI #ML #students #share #like #PMRF #IITM #MSU Prime Minister's Research Fellows (PMRF)

Here's my latest article published in Milenio about #AlphaFold 3 and its potential to disrupt medical milestones in #healthcare towards #accessibility European Bioinformatics Institute | EMBL-EBI https://lnkd.in/gYCHxpiD

📢 Last week at the 16th International Conference on Evolutionary Computation Theory and Applications (ECTA'2024) in Porto, Hugo Alcaraz-Herrera from the University of the West of England presented the work titled "Control of Biohybrid Actuators Using Neuroevolution" co-authored with Antisthenis Tsompanas, Igor Balaz, and Andy Adamatzky. The presented research proposes neuroevolution-based algorithms as the core mechanism to automatically generate controllers for biohybrid actuators that can be used on future medical devices, such as a drug-delivery catheter. 💉 ✅ The results confirmed that controllers generated by neuroevolution-based methodologies were superior in overall performance and robustness when compared with controllers generated by other methods. 📊 The presentation and the discussion session following it provided an opportunity to receive feedback from an audience of experts in a number of evolutionary computational models, including evolutionary algorithms, genetic algorithms, evolution strategy, evolutionary programming and swarm intelligence. These techniques form the basis of several disciplines such as artificial life and evolutionary robotics 🤖. Once the proceedings are announced, we will add links to the Library section of our project website 👉 https://lnkd.in/dcu5kYNt #ECTA2024 #Biohybrid #Neuroevolution #biomeld

🌌 CENTRA Research Update 🌌 José Figueiredo, #professor at Faculdade de Ciências da Universidade de Lisboa, has a new paper in Neuromorphic Computing and Engineering IOP Publishing titled "Photonic-electronic spiking neuron with multi-modal and multi-wavelength excitatory and inhibitory operation for high-speed neuromorphic sensing and computing" This study introduces a novel photonic-electronic spiking neuron built with a resonant tunneling diode with photodetection (RTD-PD), combining optical and electronic inputs for high-speed, energy-efficient neuromorphic computing. The neuron supports multi-modal and multi-wavelength control, covering telecom bands (1310 & 1550 nm), and enables excitatory and inhibitory spiking regimes, mimicking the behavior of biological neurons. It operates at incredibly fast speeds on the nanosecond scale while consuming very low energy (∼150 µW). This innovative approach enhances the potential for high-speed sensing, parallel processing, and artificial intelligence hardware, paving the way for advanced neuromorphic systems that merge photonic and electronic domains. 🔗 Explore the full study: https://lnkd.in/dGDfiqzb #NeuromorphicComputing #Photonics #ArtificialIntelligence #RTD #HighSpeedProcessing #EnergyEfficiency #ScientificInnovation #NeuromorphicSensors #FCUL

During last 24 hours, I took an interest in the work of one of the Clarivate citation laureate: Kazunari Domen. My interest follows as I (Reginald B. Little, RBL) research similar phenomena: catalysis of splitting water to hydrogen and oxygen. I (RBL) reasoned internally nuclei by nuclear and nucleon orbital angular momenta and spin angular momenta [nuclear magnetic moments (NMMs)] can couple also to surrounding electrons of such catalysts and even nonferrocatalysts and various substrates for altering electronic orbitals for novel catalysis and chemical reaction dynamics and enzymatics with first proposing in 2002 bare protons and bare neutrons manifesting such novel ferrochemistry. Thereby I first proposed in 2002 the quantum mechanical coupling of electronic wavefunction to nucleon wavefunctions in nuclei under some conditions. Then in 2004-05 I successfully tested such at National High Magnetic Field Laboratory (NHMFL) by proposing novel electrochemistry occuring inside the huge powerful magnets during operation as I proposed the copper (positive NMM) and silver (negative NMM) coils of the magnet decomposing water in stronger magnetic fields by higher currents through the coils with measurements supporting my theory of such { https://lnkd.in/ebZNc85 }. I note here my discovery in 2004-05 at NHMFL gives theoretical framework for a possibly Nobel Prize in Chemistry based on experimental work of Kazunari Domen as the recent such experimental work of Domen in 2019 made a quantum leap of in efficiency of solar radiation (sunlight) converting water to hydrogen and oxygen (water decomposition) based on use of electrodes that follows RBL's theory (2005) of positive and negative NMMs (2005). Consider the published literature, PLEASE! during the 1970s Domen as undergraduate researched: “H2O synthesis from H2 and O2 on Cu metal catalyst” at Univ of Tokyo. Then Domen earned PhD in 1982 at Univ of Tokyo exploring photocatalyst. He has been studying photocatalytic water splitting every since then. In general, during 1980s Domen made very slight advancement finding catalysts but the NMMs were more so negative NMMs. During 1990s not much advancement occurred by Domen as the explored catalyst had more so positive NMMs. During such 20 year period, Domen in 2000 noted mechanocatalysis for the mechanism. But in 2000-03, RBL proposed a mechanism for enhancing by + and - NMMs. But after 2006, large increases in quantum efficiencies occurred and this is based on Domen's subsequent use of more catalysts having more of both positive and negative NMMs. Such 2006 advancement of Domen was after RBL's theory (2004-05) of both positive and negative NMMs for catalytic decomposition of water. See comments below for details. For experimental work, I agree with Clarivate that Domen deserves Nobel Prize. BUT THERE ARE ALSO WORTHY THEORETICAL PUBLISHED DOCUMENTS, even if they are not highly cited!

I am pleased to be listed again among the top 2% of academic scientists based on Scopus citation data, ranking 50th in Biomedical Engineering (excluding self-citations single year data). While I don’t see this as a reason for overhype, such rankings highlight the impact of research on global academic peers. The "Top 2% Scientists in Biomedical Engineering - 2024" ranking reveals key leaders, institutions, and countries within the field, using metrics like the h-index, co-authorship adjusted hm-index, and composite c-score. Robert Langer (MIT), Amir A. Zadpoor (Delft University of Technology), and Dietmar W. Hutmacher (Queensland University of Technology) are the top three in Biomedical Engineering based on citation impact. The United States leads with 54 scientists, followed by Australia and China with 6 and 5, among top 100 in Biomedical Engineering, respectively. While citation metrics provide useful benchmarks, they also have limitations, including the risk of misinterpretation and the challenge of capturing qualitative contributions. Accurate Scopus profiles are essential for reliable data. This ranking encourages academics to use citation metrics wisely, emphasizing the need to consider both quantitative and qualitative aspects of scientific contributions. For ethical practices in research evaluation, the Leiden Manifesto serves as a guideline for the responsible use of these metrics. Please see the full list here https://lnkd.in/gFYqFzsf

The UQ IITD Research Academy is glad to share the first PhD journal article publication of our UQ IITD PhD scholar, Jijo Derick Abraham entitled as "Computational investigation of the role of ventricular remodelling in HFpEF: The key to phenotype dissection" in the Computers in Biology and Medicine, Elsevier (Read more: https://lnkd.in/gVM_fKMT). Key highlights from this paper includes: ▶ First computational study to address the recent paradigm of HFpEF phenotype dissection based on EF. ▶ Computationally illustrates the effect of various left ventricle (LV) remodelling patterns on LV functional indices. ▶ Computationally replicates the key features of ‘HFpEF with higher EF’ and ‘HFpEF with lower EF’. ▶ Predicts the reason for the paradoxical rise of LV chamber stiffness exponent (β) in HFpEF with higher EF even in the absence of significant fibrosis. ▶ Predicts why inotropes did not improve cardiac performance in HFpEF with higher EF. Jijo Derick Abraham is availing his PhD research on “Quantifying the mechanical properties of myocardium in cases of heart failure by preserved ejection fraction (HFpEF)” under the joint supervision of Assistant Professor Bahni Ray, Department of Mechanical Engineering, IITD and Associate Professor Christopher Leonardi, Director of Teaching and Learning, School of Mechanical and Mining Engineering, UQ and Dr. Travis Mitchell, School of Mechanical and Mining Engineering, UQ under the UQ IITD Research Academy. The research is broadly covered under the Healthcare and Biotechnology thematic area. #HFpEF #cardiacmechanics #heartfailure #cardiovascularmechanics #biomechanics #finiteelementmodelling #PhDScholars #Achievements #HDR #UQ #IITD #ResearchAcademy #HealthcareandBiotechnology

"Researchers at the University of Chicago Pritzker School of Molecular Engineering (PME) have made unexpected progress toward developing a new optical memory that can quickly and energy-efficiently store and access computational data. While studying a complex material composed of manganese, bismuth and tellurium (MnBi2Te4), the researchers realized that the material's magnetic properties changed quickly and easily in response to light. This means that a laser could be used to encode information within the magnetic states of MnBi2Te4." #materialscience #magneticmemory

I wanted to highlight a nice article written by USC Viterbi that highlights our miniaturization of pulse-mode Electron Paramagnetic Resonance (EPR) spectrometer technology on a chip. This project, led by my Ph.D. student, Ray Sun, realized the very first fully-integrated, portable pulse EPR spectrometer, which is several orders of magnitude smaller in size, power consumption, and cost than traditional EPR spectrometer lab equipment, while maintaining comparable sensitivity. EPR spectrometry is considered the gold standard for studying species with unpaired electrons and has wide-ranging applications from the biological sciences to quantum sensing and computing. The work was originally presented at the IEEE International Solid-State Circuits Conference (ISSCC) 2024 and recently extended to a journal paper and published in the IEEE Transactions on Biomedical Circuits and Systems (TBIOCAS). https://lnkd.in/g2nCzhp8 #EPR #MRI #CMOS #circuits #analog #RFIC #magneticresonance #ic #integratedcircuits