DNAcycP: a deep learning tool for DNA cyclizability prediction
Measuring intrinsic DNA cyclizability/bendability experimentally (via loop-seq) is laborious; a fast computational predictor was needed.
Papers
510 of 510 verified cases.
Automation of surgical skill assessment using a three-stage machine learning algorithm
Manual assessment of surgeons' skills is time-consuming, subjective and prone to inconsistent interpretation.
A First Step Towards Automated Species Recognition from Camera Trap Images of Mammals Using AI in a European Temperate Forest
Biology & Life Sciences
Advanced Neural Network Applications
partially solved
importance 4/10
scale stated
Manually labeling camera trap images of mammals for wildlife monitoring is slow and hard to scale; the paper automates species identification in Białowieża Forest, Poland.
Data-driven predictions of a multiscale Lorenz 96 chaotic system using machine-learning methods: reservoir computing, artificial neural network, and long short-term memory network
Climate & Environment
Neural Networks and Reservoir Computing
partially solved
importance 4/10
scale stated
Predicting the short-term evolution and long-term statistics of a multiscale chaotic Lorenz 96 system when only the large-scale variables are observable and the smaller-scale processes are never known.
Improving dysarthric speech recognition using empirical mode decomposition and convolutional neural network
Automatic speech recognizers perform poorly on dysarthric speech, making voice technology unreliable for affected speakers.
DeepConv-DTI: Prediction of drug-target interactions via deep learning with convolution on protein sequences
Biology & Life Sciences
Computational Drug Discovery Methods
ai assisted
importance 4/10
scale stated
Identifying drug-target interactions experimentally is costly and labor-intensive, and conventional protein descriptors were insufficiently informative for accurate in silico prediction.
Optimization of Molecules via Deep Reinforcement Learning
Biology & Life Sciences
Computational Drug Discovery Methods
partially solved
importance 4/10
scale stated
Optimizing molecules for drug discovery, including maximizing drug-likeness while maintaining similarity to an original molecule during medicinal chemistry lead optimization.
Insights and approaches using deep learning to classify wildlife
Biology & Life Sciences
Advanced Neural Network Applications
partially solved
importance 4/10
scale stated
Conservationists need to automatically identify and classify wildlife species in large camera-trap image datasets, which is slow and error-prone manually.
Classification of gravitational-wave glitches via dictionary learning
Physics & Astronomy
Pulsars and Gravitational Waves Research
partially solved
importance 4/10
scale stated
Gravitational-wave interferometers like advanced LIGO produce transient noise signals (glitches) that must be classified and separated from real astrophysical signals.
Toward Data‐Driven Weather and Climate Forecasting: Approximating a Simple General Circulation Model With Deep Learning
Climate & Environment
Meteorological Phenomena and Simulations
partially solved
importance 4/10
scale stated
Whether deep learning can emulate the dynamics of weather and climate general circulation models to enable data-driven forecasting.
Prediction of 8-state protein secondary structures by a novel deep learning architecture
Biology & Life Sciences
Protein Structure and Dynamics
partially solved
importance 4/10
scale stated
Accurately predicting 8-state protein secondary structure, which serves as an information bridge between primary sequence and tertiary structure for structure-based analysis.
UNSUPERVISED CLUSTERING OF TYPE II SUPERNOVA LIGHT CURVES
Astronomers needed a systematic, unbiased way to classify Type II supernova light curves ahead of upcoming large transient surveys.
Research on tire crack detection using image deep learning method
Engineering & Materials
Industrial Vision Systems and Defect Detection
solved
importance 4/10
scale stated
Tire defects such as cracks and debris are hard to detect manually, creating safety risks and labor costs for drivers and manufacturers.
Deep learning acceleration of multiscale superresolution localization photoacoustic imaging
Superresolution photoacoustic imaging achieves very high spatial resolution but requires superimposing many raw image frames, sacrificing temporal resolution and making imaging slow.
Automated claustrum segmentation in human brain MRI using deep learning
Medicine & Health
Medical Image Segmentation Techniques
partially solved
importance 4/10
scale stated
The claustrum's delicate, sheet-like structure made it infeasible to segment with conventional methods in human brain MRI, limiting in vivo research.
Deep reinforcement learning for multi-agent coordination
Robotics & Autonomy
Reinforcement Learning in Robotics
partially solved
importance 4/10
scale stated
Multiple robots working in narrow, confined environments suffer congestion and interference that hinder collective task performance, and coordinating them without explicit communication is hard.
A deep learning-based stripe self-correction method for stitched microscopic images
Stitched fluorescence microscope images contain stripes and artifacts from optics or specimens that degrade image quality and ruin downstream quantitative analysis.
Deep Learning Combinatorial Models for Intelligent Supply Chain Demand Forecasting
Business & Industry
Forecasting Techniques and Applications
ai assisted
importance 4/10
scale stated
New energy vehicle enterprises need accurate demand forecasting as a decision-making basis.
Fast Inverse Design of Transonic Airfoils by Combining Deep Learning and Efficient Global Optimization
Engineering & Materials
Computational Fluid Dynamics and Aerodynamics
solved
importance 4/10
scale stated
Inverse design of transonic airfoils—finding a shape that produces a prescribed pressure distribution—is computationally slow.
Machine Learning on a Robotic Platform for the Design of Polymer–Protein Hybrids
Engineering & Materials
Machine Learning in Materials Science
partially solved
importance 4/10
scale stated
Designing synthetic copolymers that stabilize proteins is difficult because the chemical and composition space is too vast for rational design.
Early yield prediction in different grapevine varieties using computer vision and machine learning
Predicting grapevine yield early in the season across different varieties to support wine industry yield assessment.
Particle reconstruction of volumetric particle image velocimetry with the strategy of machine learning
3D particle reconstruction from limited 2D projections in volumetric PIV is an under-determined inverse problem that iterative methods solve only approximately.
A Simulation-Data-Based Machine Learning Model for Predicting Basic Parameter Settings of the Plasticizing Process in Injection Molding
Engineering & Materials
Injection Molding Process and Properties
partially solved
importance 4/10
scale stated
Finding optimal machine settings for the plasticizing process in injection molding normally requires time-consuming, expensive physical trials.
In-situ identification of material batches using machine learning for machining operations
Engineering & Materials
Advanced machining processes and optimization
partially solved
importance 4/10
scale stated
Batches of the same specified material differ in machinability, and unidentified differences reduce product quality and raise manufacturing costs.