Around 20% of patients with non-alcoholic fatty liver disease (NAFLD) are lean. Increasing evidence suggests that lean NAFLD is a unique subtype of the disease. We aimed to explore the metabolic profile, genetic basis, causal risk factors, and clinical sequelae underlying lean NAFLD. NAFLD was diagnosed by whole liver proton density fat fraction ≥5%. Whole liver proton density fat fraction and hepatic iron were quantified using magnetic resonance imaging in the UK Biobank. Individuals in this study were stratified according to the World Health Organization criteria of obesity, into lean, overweight, and obese. Mediation analysis, Mendelian randomisation analysis, and Bayesian networks were used to identify a risk factor or a clinical sequela of lean/obese NAFLD. Lean NAFLD manifested a distinct metabolic profile, featured by elevated hepatic iron and fasting glucose. Four loci, namely, HFE rs1800562, SLC17A3-SLC17A2-TRIM38 rs9348697, PNPLA3 rs738409, and TM6SF2 rs58542926, were associated with lean NAFLD (p <5 × 10^-8). HFE rs1800562 was specifically associated with lean NAFLD and demonstrated a significant mediation effect through elevating hepatic iron. Type 2 diabetes was the most pronounced clinical sequela of lean NAFLD, followed by liver cirrhosis. Our study suggested that HFE plays a potential steatogenic role rather than regulating iron homoeostasis in patients with lean NAFLD. The increased liver iron deposition is associated with lean NAFLD, whereas obese NAFLD is not related to hepatic iron. The clinical management of patients with lean NAFLD shall be concerned with the prevention and treatment of type 2 diabetes and liver cirrhosis. Lean NAFLD has a distinct natural history from obese NAFLD. This study underscored liver iron content and the genetic variant of the iron homoeostasis gene HFE as major risks of lean NAFLD, in addition to the unique metabolic profile. The development of type 2 diabetes or liver cirrhosis shall be closely monitored and prevented in patients with lean NAFLD.

As a key task in autonomous driving, 3D object detection based on LiDAR-camera fusion is expected to achieve more robust results by the complementarity of the two sensors. However, LiDAR-camera fusion is non-trivial. An existing problem for this type of detector is that the scale and receptive field of LiDAR point features and image features are not matched, leading to information deficiency or redundancy in fusion. This paper proposes a Point-based Pyramid Attention Fusion (PPAF) module for LiDAR-camera fusion to solve the problem. The PPAF module learns corresponding image features of LiDAR points with a matched scale based on the image feature pyramid and attention mechanism for a better effect of fusion. Furthermore, based on the PPAF module, a new LiDAR-camera fusion-based 3D object detector named FS-Net is proposed, a two-stage detector with LiDAR voxel-based RPN and refinement network based on enriched LiDAR-camera features. Experiments on two public datasets demonstrate the effectiveness of our approach.

The existing commodity Wi-Fi-based human gait recognition systems mainly focus on a single subject due to the challenges of multisubject walking monitoring. To tackle the problem, we propose Wi-Diag, the first commodity Wi-Fi-based multisubject abnormal gait diagnosis system that leverages only one pair of off-the-shelf commercial Wi-Fi transceivers to separate each subject’s gait information and maintains an excellent performance when the scenario changes. It is an intelligent multisubject gait diagnosis system that can release an experienced doctor from heavy load work. Multisubject abnormal gait diagnosis is modeled as a blind source separation (BSS) issue, and multisubject walking mixed signals are efficiently separated by IC analysis (ICA) approach. This fact is verified by comprehensive theoretical derivation and experimental validation. In addition, CycleGAN is leveraged to mitigate the environmental dependency so that Wi-Diag can be robust when the scenario changes. The excellent performance of Wi-Diag is verified by extensive experiments. The average mean diagnosis accuracy with a maximum group size of four and various scenarios is 87.77%.

Currently, a novel of meta-heuristic algorithm called monarch butterfly optimization (MBO) is presented for solving machine learning and continuous optimization problems. It has been proved experimentally that MBO is superior to artificial bee colony algorithm (ABC), ant colony optimization algorithm (ACO), Biogeography-based optimization (BBO), differential evolution algorithm (DE) and simple genetic algorithm (SGA) algorithms on most test functions. This paper presents a new version of MBO with simulated annealing (SA) strategy called SAMBO. The SA strategy is put in the migration operator and butterfly adjusting operator. So the newly proposed algorithm has two features: One is that the algorithm accepts all the butterfly individuals whose fitness are better than their parents. The other is that the algorithm randomly selects some individuals which are worse than their parents to disturbance the convergence of algorithm. In this way, the SAMBO algorithm can escape from local optima. Finally, the experiments are carried on 14 continuous nonlinear functions, and results show that SAMBO method exceeds the MBO algorithm on most test functions.