Feed aggregator

A novel approach to the maxillary nerve: The Palatine Technique

Iliana graduated from the University of Thessaloniki in 2017. After undertaking 4 internships and spending a couple of years in first opinion practice, she decided to pursue her dream of becoming a veterinary anaesthesiologist. She started her residency at the University of Cambridge in 2023.

Chaired by Muriel Dresen and Olivier Restif

• Speaker: Iliana Antonopoulou , Department of Veterinary Medicine
• Friday 02 May 2025, 08:45-10:00
• Venue: LT2.
• Series: Friday Morning Seminars, Dept of Veterinary Medicine; organiser: Fiona Roby.

Add to your calendar or Include in your list

Piezoelectric materials with high piezoelectric coefficient (d 33) and high mechanical quality factor (Q m) are critical for advanced high-power applications. However, achieving this combination is challenging due to the inherent trade-off between d 33 and Q m. This work reports a novel strategy involving the introduction of defect dipoles into quadruple point compositions to break the conventional trade-off and simultaneously enhance d 33 to 710 pC/N and Q m to 929 in designed lead-free piezoelectric materials.

Abstract

Piezoelectric materials with a high piezoelectric coefficient (d 33) and high mechanical quality factor (Q m) are vital for advanced high-power applications. However, achieving this combination is challenging, particularly for lead-free piezoelectrics, because a high d 33 value relies on mobile domain walls, which increase dissipative losses and reduce Q m. In this study, this longstanding trade-off is overcome by introducing defect dipoles (via Mn doping) into the quadruple point (QP) composition of the lead-free Ba(Sn, Ti)O3 system. The resultant 0.5%Mn-doped Ba(Sn0.11Ti0.89)O3 (BST-0.5%Mn) ceramic exhibits a high d 33 value of 710 pC/N and high Q m value of 929, while the BST-1%Mn ceramic achieves a d 33 value of 614 pC/N and Q m value of 1138. These values represent a 10-fold increase in Q m and 1.6-fold increase in d 33 for BST-0.5%Mn, compared to those for undoped BST. High-resolution scanning transmission electron microscopy and phase-field simulations reveal that the enhanced d 33 and Q m are attributable to the coexistence of multiple phases of QPs with symmetry-conforming defect dipoles, challenging the long-held notion of physical incompatibility between high d 33 and high Q m. These findings offer a pathway for designing eco-friendly piezoelectric materials with unprecedented performance, paving the way for sustainable and efficient high-power applications.

A hybrid lipoplex (RMC/pApoE2) composed of guanidine-rich lipids (metformin-inspired MLS and arginine-contained RLS), oleic acid-modified cerium dioxide (OA@CeO2), and pApoE2 restores neuron-microglia crosstalk to eliminate Aβ deposits from temporal and spatial multidimensions. It induces neuronal and microglial autophagy at the upstream and downstream of Aβ metabolism, along with enhanced pApoE2 transfection for intra- and extra-cellular elimination of Aβ aggregates.

Abstract

Efficient clearance of amyloid-β (Aβ) is vital but challenging in Alzheimer's disease (AD) treatment due to its complicated regulation mechanisms during generation and metabolism. It necessitates a multidimensional synergistic strategy based on ingenious delivery system design. Herein, guanidine-rich lipids (metformin-inspired MLS and arginine-contained RLS) are devised to trigger selective chaperone-mediated autophagy for amyloid precursor protein degradation in neurons. They are further co-assembled with oleic acid-modified cerium dioxide (OA@CeO2) to form RMC assembly for pApoE2 delivery (RMC/pApoE2 lipoplex). The OA@CeO2 boosts macro-autophagy, alleviates oxidative stress and inflammatory microenvironment, and promotes the neurons-microglia crosstalk for Aβ elimination. Concurrently, both guanidine-rich lipids and OA@CeO2 benefit pApoE2 transfection in neurons, enabling the transport of Aβ into microglia, and facilitating enzymatic hydrolysis and cellular digestion of extracellular Aβ. The lipoplex-boosted neuron–microglia interactions ultimately eliminate both intra- and extra-cellular Aβ aggregates. Consequently, the RMC/pApoE2 lipoplex eliminates ≈86.9% of Aβ plaques in the hippocampus of APP/PS1 mice and restored the synaptic function and neuronal connectivity. Moreover, it recovers the spatial memory of APP/PS1 mice to nearly the level of WT control. The presented hybrid lipoplex showcases an advanced gene delivery system, and offers a promising strategy for Aβ clearance in AD treatment.

A self-organized nanocomposite addresses the challenge of energy storage materials degrading at high temperatures. The nanocomposite maintains exceptional performance from room temperature to 200 °C, with enhanced energy density and efficiency. Through the trirelaxor matrix and nanointerface with antiferroelectric nanoparticles, it offers a promising strategy for high-temperature dielectric energy storage, overcoming polarization and breakdown degradation at elevated temperatures.

Abstract

A fundamental paradox in energy storage dielectrics lies in the challenge of achieving superior performance consistently across both room and elevated temperatures. This is addressed by designing a self-organized nanocomposite (1−x)(Ba,Sr)(Ti,Sn)O3-xBi1.5ZnNb1.5O7 composed of nano-sized antiferroelectric(AFE) particles embedded into a trirelaxor(TRE) matrix through nanoscale phase separation process. The optimal composition at x = 0.11 exhibits outstanding energy storage performance from room temperature (energy density = 8.5 J cm−3, efficiency = 94.8%, and figure of merit of 167 J cm−3) up to 200 °C (energy density = 4.85 J cm−3, efficiency >90% and figure of merit of 49 J cm−3), outperforming existing Pb-free dielectrics. High-resolution transmission electron microscopy and synchrotron x-ray diffractometry reveal that the coexisting nanometric antiferroelectric particles and the trirelaxor nanodomains sustain over a wide temperature range. Piezoresponse force microscopy and phase-field simulation show that hysteresis-free switching of trirelaxor nanodomains enables enhanced polarization and low hysteretic loss. Resistivity shows a 2–3 order of magnitude increases accompanying significant increase in breakdown strength up to high temperatures, attributable to deep charge trapping effect at high-density TRE/AFE interfaces as evidenced by thermally stimulated depolarization current. These favorable effects in the nano-composite are responsible for its high energy storage performance up to high temperatures.

A cost-effective symmetric PbSe-based device constructed from seven pairs of Pb0.988Cu0.002Se (p-type) and Pb1.02Cu0.002Se (n-type), which demonstrates impressive cooling temperature difference (ΔT C) of 32.8, 36.9, and 41.0 K with the hot side maintained at 303, 323, and 343 K, respectively.

Abstract

Thermoelectric cooling technology has broad applications but is limited by the high cost of tellurium (Te) in commercially available Bi2Te3-based thermoelectric materials. Herein, a cost-effective symmetric PbSe-based device constructed from 7 pairs of Pb0.988Cu0.002Se (p-type) and Pb1.02Cu0.002Se (n-type) is presented, which demonstrates impressive cooling temperature difference (ΔT C) of 32.8 and 41.0 K with the hot side maintained at 303 and 343 K, respectively. This low-cost symmetric PbSe-based device exhibits superior cost-effectiveness (ΔT/cost) for near-room-temperature thermoelectric cooling compared to other Bi2Te3-based devices. Its high cooling performance primarily stems from an advanced carrier and phonon transport properties in p-type Pb0.988Cu0.002Se. Specifically, Pb vacancy and Cu substitution in Pb0.988Cu0.002Se act as strong p-type dopants that effectively optimize carrier density, resulting in a maximum power factor of 28.69 µW cm−1 K−2 at room temperature. Moreover, the mobile Cu atoms within the lattice significantly impede phonon propagation, leading to a low room-temperature lattice thermal conductivity of 1.10 W m−1 K−1. Finally, the room-temperature figure of merit (ZT) and average ZT value in p-type Pb0.988Cu0.002Se can reach 0.6 and 0.68 at 300–573 K, surpassing previous p-type PbSe-based polycrystals. This work emphasizes the significant potential of a cost-effective PbSe compound for near-room-temperature cooling applications.

An atomically ordered, spin-polarized Co-doped PdCu intermetallic compound is rationally synthesized. The introduction of spin-polarized Co atoms can enhance the *NO binding and hydrogenation on its N-side to form *HNO, which further produces *NH2OH. The subsequent coupling of *CO and *NH2OH leads to the efficient and stable formation of urea.

Abstract

The electrocatalytic conversion of NO3 − and CO2 into urea features a potential means of reducing carbon footprint and generating value-added chemicals. Nonetheless, due to the limited efficiency of carbon−nitrogen (C─N) coupling and the competing side reaction that forms ammonia, the urea selectivity and production yield have remained low. In this work, a spin−polarized cobalt−doped, atomically ordered PdCu intermetallic compound (denoted as PdCuCo) is developed as an efficient urea electrosynthesis catalyst. The Pd and Cu serve as the adsorption sites for CO2 and NO3 −, respectively, and the spin−polarized Co sites promote the adsorption of *NO intermediate, followed by hydrogenation of *NO at its N−terminal to form *HNO, instead of at its O−terminal. The difference in the hydrogenation position switches the subsequent reaction pathway to produce urea, in contrast to the PdCu or Ni−doped PdCu intermetallic compounds with main product selectivity of ammonia. The PdCuCo electrocatalyst exhibited an outstanding electrosynthesis of urea from NO3 − and CO2, including a Faradaic efficiency of 81%, a high urea yield of 227 mmol gcat. −1 h−1, and a notable electrochemical stability of >260 h, suggesting the attractive potential of designing spin−polarized catalytic sites for carbon−nitrogen coupling processes.

Novel guidance conduits are fabricated through electrospinning and masked coaxial electrospraying, integrating topological, haptotactic, and chemotactic cues to promote cell migration, neural stem cell differentiation, and axonal extension. In rat models, these conduits inhibited fibroblast proliferation, preserved microglial homeostasis, and promoted neuronal regeneration, significantly improving functional recovery and offering a promising strategy for spinal cord injury treatment.

Abstract

Spinal cord injury (SCI) is a debilitating condition that leads to severe disabilities and imposes significant economic and social burdens. Current therapeutic strategies primarily focus on symptom management, with limited success in promoting full neurological recovery. In response to this challenge, the design of novel guidance conduits incorporating multiple gradient cues, inspired is reported by biological processes, to enhance spinal cord repair. These conduits are fabricated using electrospinning and masked coaxial electrospraying, a simple yet effective method that integrates topological, haptotactic, and chemotactic cues into a single scaffold. The synergy of these cues significantly promoted cell migration, neural stem cell differentiation into neurons, and axonal extension, resulting in substantial improvements in spinal cord regeneration and functional recovery in a rat model. Single-nucleus RNA sequencing further demonstrated that the guidance conduit inhibited fibroblast proliferation, preserved microglial homeostasis, restored cellular proportions, and facilitated the regeneration of neuronal axons, dendrites, and synapses. This work presents an innovative, versatile platform for fabricating tissue scaffolds that integrate multiple gradient cues, offering a promising strategy for SCI treatment and broader tissue regeneration applications.

This research, highlighting i) the hierarchical assembly of Photonins, ii) the sugar–lectin binding which modulates structural stability and mechanical properties of sea silk, and iii) the use of golden sea silk as a photonic protein fiber.

Abstract

The harvesting of sea silk, a luxurious golden textile traditionally obtained from the endangered mollusk Pinna nobilis, faces severe limitations due to conservation efforts, driving the search for sustainable alternatives. Atrina pectinata, a phylogenetically close relative within the Pinnidae family is identified, as a viable source of biomimetic sea silk. The byssal threads of A. pectinata can be processed using existing methods, providing a way to continue producing this historically significant textile. These threads exhibit a remarkable hierarchical structure with globular proteins organized across multiple scales and stabilized by supramolecular sugar-lectin interactions that influence their mechanical properties. Moreover, the threads display a brilliant golden hue arising from structural coloration, ensuring exceptional lightfastness, retaining their color for millennia. This discovery elucidates the biomolecular foundations of sea silk's unique properties and establishes A. pectinata as a sustainable candidate for producing exquisite golden textiles and bioinspired pigments, thereby addressing the growing demand for eco-friendly and long-lasting colored materials in the textile and pigment industries.

Achieving synergistic strengthening and toughening of natural fiber-reinforced composites remains a significant challenge. Drawing inspiration from multi-layer helical structures observed in natural organisms, a green, facile, and versatile transitional unit design strategy is designed to construct a gradual helical structure. This approach helps successfully achieve simultaneous strengthening and toughening of natural fiber-reinforced composites.

Abstract

Multilayered helical arrangements are commonly observed in natural creatures to enhance their strength and toughness. A biomimicry of such an intricate structure has thus far been challenging. Herein, a green, facile, and versatile design strategy is proposed for transitional units. The proposed strategy is applied to develop a gradual helical (GH) structure that can reinforce thermoplastics using bamboo fibers (≈20 cm). A transitional unit is constructed through a combination of rolling and twisting. Following hot pressing, a biomimetic fiber-reinforced composite with a GH structure is fabricated. The GH structure is made up of 3D helical fibers with a gradual variation in the helical angle from the surface to the core, achieving minimal staggered angles and bridging of different fiber layers. Owing to stress decomposition and transfer as well as the coupling effect of the helical fibers, the GH structure exhibits outstanding tensile and bending strengths. Moreover, owing to the staggered arrangement, bridging, and deformation behavior of the fibers, the GH structure achieves remarkable impact toughness through crack deflection and fiber uncoiling. The GH structure and transitional unit assembly strategy can facilitate the development of advanced composites with superior mechanical properties through an environmentally friendly, simple, and versatile structural design approach.

This work introduces an efficient dual-directional upcycling scheme enabled through a mechanical homogenization pretreatment. It enables various layered oxide cathodes to be reprocessed into fresh NCM cathodes with tailored Ni contents through boosted atomic diffusion in just 4 h of solid-state sintering. Delivering upcycled cathodes with comparable electrochemical performances to their commercial counterparts, this approach excels itself from the cost-effectiveness over conventional acid-leaching resynthesis approaches.

Abstract

Upcycling is regarded as a sustainable and promising recycling solution for spent lithium-ion batteries (LIBs). However, current upcycling strategies such as converting Ni-lean to Ni-rich cathodes struggle to change the composition of the spent cathodes to meet the diverse market demands. In addition, the commonly employed molten-salts method requires tens of hours of high-temperature treatment, restricting its sustainability. Herein, this study reports an efficient, flexible dual-directional upcycling strategy to upcycle a broad family of layered oxide cathodes into fresh LiNixCoyMnzO2 (NCM) cathodes with tailored Ni-contents—either increased or decreased—in just 4 h via mechanical homogenization pretreatment. This study confirms that the bulk diffusion of transition metals (TMs) is the rate-determining step in the resynthesis process, and the mechanical homogenization can shorten the diffusion pathway of TMs, thus reducing the sintering duration effectively. The as-upcycled NCM cathodes can deliver electrochemical performance on par with commercial counterparts. Notably, a systematic technoeconomic analysis shows that upcycling spent LiCoO2 into NCM622 can yield a profit up to 35 US$/kg, 30% higher than the conventional acid-leaching resynthesis approach. This work provides an energy-saving, widely adaptable, flexible, and cost-efficient method for regenerating spent cathode materials, paving the way for the sustainable recycling of LIBs.

A multifunctional memristor is demonstrated for in-memory sensing and computing, leveraging a MoWS₂/VOx heterojunction to enable high ON/OFF ratio up to 10⁸ with ultralow operating voltages of ±0.2 V. This bio-inspired multimodal design exhibits tunable synaptic behavior across electrical, optical, and humidity stimuli, enabling in situ modulation of conductance for low-power, real-time processing of multisensory signals. The reconfigurable humidity-adaptive neuron and humidity-mediated optical synaptic learning enable non-contact respiratory sensing and vision clarity control, paving the way for energy-efficient next-generation human–machine interfaces.

Abstract

Advancements in computing have progressed from near-sensor to in-sensor computing, culminating in the development of multimodal in-memory computing, which enables faster, energy-efficient data processing by performing computations directly within the memory devices. A bio-inspired multimodal in-memory computing system capable of performing real-time low power processing of multisensory signals, lowering data conversion and transmission across several modules in conventional chips is introduced. A novel Cu/MoWS2/VO x /Pt based multimodal memristor is characterized by an ON/OFF ratio as high as 108 with consistent and ultralow operating voltages of ±0.2 surpassing conventional single-mode memory functions. Apart from observing electrical synaptic behavior, photonic depression and humidity mediated optical synaptic learning is also demonstrated. The heterojunction with MoWS2 also enables reconfigurable modulation in both memory and optical synaptic functionalities with changing humidity. This behavior provides tunable conductance modulation capabilities emulating synaptic transmission in biological neurons while showing potential in respiratory detection module for healthcare application. The humidity sensing capability is implemented to demonstrate vision clarity using a convolutional neural network (CNN), with different humidity levels applied as a data augmentation preprocessing method. This proposed multimodal functionality represents a novel platform for developing artificial sensory neurons, with significant implications for non-contact human–computer interaction in intelligent systems.

Poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) based conductive hydrogels have received great attention in bioelectronics on account of their tissue-like mechanical properties. However, inhomogeneous morphologies of the conducting PEDOT phase limits their electrical and mechanical properties. Here, supramolecular hydrogels with self-doped PEDOT (S-PEDOT) homogeneously distributed are reported, which simultaneously exhibit high toughness (620 kJ m−3), softness (10.5 kPa) and conductivity (5.8 S cm−1).

Abstract

Mechanically resilient hydrogels with ion-electron mixed transport properties effectively bridge biology with electronics. An ideal bioelectronic interface can be realized through introducing electronically conductive polymers into supramolecular hydrogels. However, inhomogeneous morphologies of conducting polymers, such as poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS), have limited mechanical properties and ion-electron interactions. Here, supramolecular conductive hydrogels that possess homogeneous ionic and electronic transport are achieved. The materials demonstrate high toughness (620 kJ m−3), stretchability (>1000%), softness (10.5 kPa), and conductivity (5.8 S cm−1), which surpasses commonly used inhomogeneous PEDOT:PSS-based hydrogels. The homogeneous network leads to higher charge injection capacitance and lower skin impedance compared to commercial electrodes or commonly used inhomogeneous PEDOT:PSS conducting networks. This significant advance arises from the homogeneous incorporation of the hydrophilic self-doped conducting polymer S-PEDOT, which has polymerized within a supramolecular polymer network template mediated by high-binding affinity host-guest crosslinks. Furthermore, the compatibility of S-PEDOT with hydrophilic secondary networks enables the realization of fully dryable and reswellable electronic devices, facilitating reusability and improving their ease of handling. It is anticipated that achieving such material architectures will offer a promising new direction in future synthesis and implementation of conductive hydrogels in the field of bioelectronics.

A bilayer electrode–electrolyte interface engineering strategy is presented to introduce sodium thioctate into bare ZnSO4 electrolytes. Benefiting from the bilayer electrode-electrolyte interface, optimized solvation structure, and reconstructed inner Helmholtz plane, the resulting Zn−MnO2 batteries exhibit prominent cycling stability. This work provides effective guidance for the rational design of safe and long-life aqueous zinc-ion batteries.

Abstract

Understanding the composition–characteristics–performance relationship of the electrolyte–electric double layer–electrode–electrolyte interface (EEI) is crucial to construct stable EEIs for high-performance aqueous Zn–MnO2 batteries (AZMBs). However, the interaction mechanisms in AZMBs remain unclear. This work introduces sodium thioctate (ST) into ZnSO4 electrolyte to construct a stable bilayer EEI on both Zn and MnO2 electrodes. First, zincophilic ST regulates the solvation structure of hydrated Zn2+, suppressing corrosion and the hydrogen evolution reaction. Second, the specific adsorption of ST reconstructs the inner Helmholtz plane, facilitating the desolvation of hydrated Zn2+ and homogenizing charge distribution. Finally, ST molecules undergo reversible polymerization at the interface, forming a stable bilayer EEI with a poly(zinc thioctate) outer layer and a ZnS–organic amorphous inner layer, which ensures uniform zinc-ion flux and enhances mechanical stability. Additionally, the dynamic disulfide bonds in ST further enable self-regulation and self-healing of the interface, mitigating damage during cycling. As a result, the ST-enhanced Zn symmetric battery achieves 7800 cycles at 60 mA cm−2, while the AZMB exhibits only 0.0014% capacity decay over 10 000 cycles at 2000 mA g−1. This bilayer EEI engineering strategy offers effective guidance for the rational design of safe and long-life aqueous zinc-ion batteries.

Nature Nanotechnology, Published online: 29 April 2025; doi:10.1038/s41565-025-01912-5

Nanoscale, covalently bonded GeSe crystals can withstand up to 12.8% recoverable tensile strain through an atomic mechanism called reversible shuffle twinning, giving rise to anisotropic superelasticity.

Nature Nanotechnology, Published online: 29 April 2025; doi:10.1038/s41565-025-01923-2

Porous and hollow silk fibroin microneedles nanofabricated with phase front assembly allow the authors to establish a prolonged interface with plants for delivery of micronutrients, crop fortification and continuous plant health monitoring.

Nature Materials, Published online: 29 April 2025; doi:10.1038/s41563-025-02228-4

By introducing an unconventional spin–orbit torque, the authors demonstrate an all-electrical way to bidirectionally switch the perpendicular chiral antiferromagnetic order in Mn3Sn, promoting the application of antiferromagnetic spintronics.

Synthesis RIG Postdoc Seminar - Dr Antti Lahdenpera and Dr Sona Krajcovicova

“Strategies for controlling enantioselectivity in radical reactions” and “Novel Synthetic Approaches for Next-Generation Therapeutics”

• Speaker: Dr Antti Lahdenpera and Dr Sona Krajcovicova
• Wednesday 07 May 2025, 14:00-15:00
• Venue: Dept. of Chemistry, Wolfson Lecture Theatre.
• Series: Synthetic Chemistry Research Interest Group; organiser: Dr. Robert J. Phipps.

Add to your calendar or Include in your list

Gravitational Wave Signatures of Dark Matter in Neutron Star Mergers

Binary neutron star mergers provide insights into strong-field gravity and the properties of ultra-dense nuclear matter. These events offer the potential to search for signatures of physics beyond the standard model, including dark matter. We present the first numerical-relativity simulations of binary neutron star mergers admixed with dark matter, based on constraint-solved initial data. Modeling dark matter as a non-interacting fermionic gas, we investigate the impact of varying dark matter fractions and particle masses on the merger dynamics, ejecta mass, post-merger remnant properties, and the emitted gravitational waves. Our simulations suggest that the dark matter morphology – a dense core or a diluted halo – may alter the merger outcome. Scenarios with a dark matter core tend to exhibit a higher probability of prompt collapse, while those with a dark matter halo develop a common envelope, embedding the whole binary. Furthermore, gravitational wave signals from mergers with dark matter halo configurations exhibit significant deviations from standard models when the tidal deformability is calculated in a two-fluid framework neglecting the dilute and extended nature of the halo. This highlights the need for refined models in calculating the tidal deformability when considering mergers with extended dark matter structures. These initial results provide a basis for further exploration of dark matter’s role in binary neutron star mergers and their associated gravitational wave emission and can serve as a benchmark for future observations from advanced detectors and multi-messenger astrophysics.

• Speaker: Violetta Sagun, University of Southampton
• Friday 30 May 2025, 13:00-14:00
• Venue: Potter room/Zoom.
• Series: DAMTP Friday GR Seminar; organiser: Xi Tong.

Add to your calendar or Include in your list

Title to be confirmed

Abstract not available

• Speaker: Giovanna Maria Dimitri, University of Siena (Italy)
• Tuesday 27 May 2025, 13:00-14:00
• Venue: Lecture Theatre 2, Computer Laboratory, William Gates Building.
• Series: Artificial Intelligence Research Group Talks (Computer Laboratory); organiser: Mateja Jamnik.

Add to your calendar or Include in your list

Explainable AI in Neuroscience: From Interpretability to Biomarker Discovery

Explainability plays a pivotal role in building trust and fostering the adoption of artificial intelligence (AI) in healthcare, particularly in high-stakes domains like neuroscience where decisions directly affect patient outcomes. While progress in AI interpretability has been substantial, there remains a lack of clear, domain-specific guidelines for constructing meaningful and clinically relevant explanations. In this talk, I will explore how explainable AI (XAI) can be effectively integrated into neuroscience applications. I will outline practical strategies for leveraging interpretability methods to uncover novel patterns in neural data, and discuss how these insights can inform the identification of emerging biomarkers. Drawing on recent developments, I will highlight adaptable XAI frameworks that enhance transparency and support data-driven discovery. To validate these concepts, I will present illustrative case studies involving large language models (LLMs) and vision transformers applied to neuroscience. These examples serve as proof of concept, showcasing how explainable AI can not only translate complex model behavior into human-understandable insights, but also support the discovery of novel patterns and potential biomarkers relevant to clinical and research applications.

• Speaker: Mike Mamalakis (University of Cambridge)
• Tuesday 13 May 2025, 13:00-14:00
• Venue: Lecture Theatre 2, Computer Laboratory, William Gates Building.
• Series: Artificial Intelligence Research Group Talks (Computer Laboratory); organiser: Mateja Jamnik.

Add to your calendar or Include in your list