Advanced Science, EarlyView.By overcoming the fixed‐path limitations of conventional machine learning, a heterogeneous graph neural network fundamentally reconstructs material data representation. Integrating variable processing sequences with intrinsic elemental features, this framework enables exploratory optimization across high‐dimensional spaces.
Jie Yin +12 more
wiley +1 more source
Optimising a Simple Fully Convolutional Network for Accurate Brain Age Prediction in the PAC 2019 Challenge. [PDF]
Front Psychiatry, 2021 Gong W +4 more
europepmc +1 more source
Efficient Screening of Organic Singlet Fission Molecules Using Graph Neural Networks
Advanced Science, EarlyView.A high‐throughput screening framework based on graph neural networks (GNNs) and multi‐level validation facilitates the identification of singlet fission (SF) candidates. By efficiently predicting excitation energies across 20 million molecules, and integrating TDDFT calculations, synthetic accessibility assessments, and GW+BSE calculations, this ...
Li Fu +5 more
wiley +1 more source
Fully convolutional network-based multi-output model for automatic segmentation of organs at risk in thorax. [PDF]
Sci Prog, 2021 Zhang J +6 more
europepmc +1 more source
Broadening Hard‐Magnet Discovery Beyond Symmetry Constraints via Unified Effective Anisotropy
Advanced Science, EarlyView.A unified effective‐anisotropy descriptor (Keff) extends hard‐magnet screening across all seven crystal systems, beyond the uniaxial restriction of conventional searches. Machine‐learning screening of 9320 known ferromagnets and diffusion‐model generation together yield 38 rare‐earth‐free or ‐lean candidates with DFT‐validated magnetic hardness (κ > 1),
Hojae Kim +5 more
wiley +1 more source
Ferroelectric Devices for In‐Memory and In‐Sensor Computing
Advanced Science, EarlyView.Inspired by biological systems, in‐memory and in‐sensor computing overcome von Neumann bottlenecks. Ferroelectric devices can mimic synaptic functions and sense stimuli like light or force, therefore are ideal for these paradigms. This review introduces the ferroelectric devices applied for in‐memory and in‐sensor computing, covering their structures ...
Hong Fang +5 more
wiley +1 more source
Delineation of ischemic lesion from brain MRI using attention gated fully convolutional network. [PDF]
Biomed Eng Lett, 2021 Karthik R +3 more
europepmc +1 more source
Advances and Perspectives in Graphene‐Based Quantum Dots Enabled Neuromorphic Devices
Advanced Science, EarlyView.Graphene‐based QDs are zero‐dimensional carbon nanomaterials with pronounced quantum confinement and tunable electronic structures. Herein, we summarize their synthesis strategies and functionalization methods, and highlight their functional roles and operating mechanisms in devices, as well as recent advances in neuromorphic electronics. We anticipate
Yulin Zhen +9 more
wiley +1 more source
Automatic detection of COVID-19 disease using U-Net architecture based fully convolutional network. [PDF]
Biomed Signal Process Control, 2021 Kalane P, Patil S, Patil BP, Sharma DP.
europepmc +1 more source
Advanced Science, EarlyView.A conversion‐resolved constitutive framework is developed for the hydrogen‐based direct reduction of iron oxide pellets. Effective reaction and transport timescales are inferred directly from measured trajectories and mapped against operating conditions, pellet architecture, and composition. The analysis reveals how late‐stage transport control emerges
Anurag Bajpai +3 more
wiley +1 more source