MOTIVATION: Understanding cell fate determination is crucial in developmental biology and regenerative medicine. Although theoretical frameworks such as epigenetic landscape and gene regulatory networks have been proposed for decades, traditional studies have often been limited by population-averaging and low-throughput techniques, which obscure the heterogeneity of individual cells and fail to provide a systematic view of cell fate control. Recent advances in single-cell technologies have provided unprecedented resolution, revealing the complexity of cell fate decisions and driving the need for more sophisticated computational methods. RESULTS: In this review, we first emphasize experimental advances, such as single-cell multi-omics, lineage tracing, and perturbation techniques, which produce novel data modalities and enable dynamic tracking of cell fate transitions. We then discuss the modeling paradigms for cell fate studies and further assess the role of emerging AI tools in perturbation modeling and discuss the potential of single-cell and spatial foundation models. Additionally, we highlight several case studies on predicting and manipulating cell fates, and discuss key challenges and future directions of the field. AVAILABILITY AND IMPLEMENTATION: This work generates no new software.

Pedestrian detection in crowded situation is a challenging task. This study presents a straightforward and effective method called Det RCNN to detect pedestrians in crowded situation, while also pairing the body and head of individual pedestrian. On the one hand, pedestrians' heads have their characteristics of stable shape and distinct feature. On the other hand, their heads are usually positioned higher in image, so even in crowded situation, it is difficult to completely cover the pedestrians' heads. Therefore, this study equipped the DETR model with a Head Decoder (HDecoder) parallel to the Decoder. HDecoder takes the head knowledge generated in the Decoder phase as head queries. Simultaneously, the HDecoder uses a key-query mechanism to search the entire image for the body bounding boxes corresponding to the head queries. Lastly, the proposed method conducts a straightforward IOU (Intersection over Union) matching between the body bounding boxes produced in the Decoder and HDecoder phases. This HDecoder resembles the second stage of the Faster RCNN model, hence this paper termed it Det RCNN (DETR RCNN). Compared to Deformable DETR, the experimental results on the CrowdHuman dataset show that the proposed model can increase AP$_{m}$from 53.02 to 53.87. Furthermore, the mMR$^{-2}$decreased from 52.46 to 42.32 compared to the existing BFJ. The code and experiments will soon be open-sourced athttps://github.com/zefeichen/Det-RCNN.

Existing arterial signal coordination predominantly adopts physical traffic performance indicators or alternative progression bandwidth as optimization objectives, which lack a constraint mechanism between traffic efficiency and the environmental impacts associated with signal parameters. This study proposes a novel arterial ecosignal coordination approach to minimize gas emissions, energy consumption, and passenger delays by leveraging vehicle trajectory data. Initially, a vehicle kinematic analytical model incorporating driving behavior is developed to extract second-by-second individual vehicle trajectories. Through the analysis of micro-trajectory data and comparison with the evaluation metrics from VISSIM simulator, the model is validated to accurately simulate vehicle operating conditions. Subsequently, an exact ecological cost unit, encompassing gas emissions, fuel consumption, and delays for all vehicles by explicitly accounting for both electric and conventional fuel vehicles, is calculated based on the trajectory data. This cost unit is then selected as the objective function of the optimization problem, which is formulated as a bi-level model. Numerical experiments on a five-intersection arterial segment demonstrate that the proposed ecosignal coordination can significantly enhance environmental benefits at the cost of a small amount of delay growth compared to existing methods.

Abstract In pedestrian detection task, numerous predicted boxes and their corresponding scores are generated and these scores are used to filter these predicted boxes by non‐maximum suppression. This paper analysed the training process of the popular anchor‐based pedestrian detection models (e.g. YOLO and Faster RCNN), and found that the score of the predicted box reflects the overlap between the corresponding anchor and the ground truth, rather than the predicted box itself. Due to the many‐to‐one strategy adopted by anchor‐based methods, multiple predicted boxes could be generated around one predicted box. This study refers to the number of other predicted boxes around the target predicted box as its local density. When a predicted box has a higher local density, it should have a greater overlap with the ground truth. Therefore, this study proposed the fused score by introducing local density into the score. The experiments showed that replacing the score with the fused score can effectively improve the model's detection accuracy. The code and experiments will soon be open‐sourced at https://github.com/zefeichen/FusedScore .

In this paper, we propose a new RGB-thermal image segmentation method based on parameter sharing and attention fusion for safe autonomous driving. An encoder-decoder network structure is adopted. The encoder, which has shared convolution layer parameters and private batch normalization layer parameters (parameter sharing scheme), is used to extract features from RGB and thermal images. The extracted features are then fused by spatial and channel attention. The output of each residual block is fused, and the self-learning weight is used to integrate the fusion information of all residual blocks of the same levels. Subsequently, the fused features are integrated through a feature integration (FI) module in the decoder. Cross-entropy supervision of segmentation and edge is performed on the outputs of the decoders. Our proposed method is evaluated and compared with 17 state-of-the-art image segmentation methods, both qualitatively and quantitatively on the MFNet dataset which includes various objects in urban scenes. The results show that the proposed method outperforms previous methods by at least 0.3% and 1.8% in MRecall and MIoU, respectively, providing foundations for the development of autonomous driving technologies for safety enhancement.

An optimization model of regional green wave coordination control for the coordinated path set is proposed based on the network’s directional and asymmetric actual coordination control demands. The model in this paper selects the coordinated path set as the optimization object and takes the best regional coordination control effect as the optimization objective. By selecting signal timing parameters, such as stage time, stage center point, offset, and cycle length, as optimization variables, the relationship between signal timing parameters and the effect of green wave coordination was established. The stage time of the coordinated stage at the upstream and downstream intersections is used to calculate the evaluation index of the coordination control benefit of the whole road network. The case study demonstrates that this model can optimize coordinated path chains with left-turn traffic flows under green wave coordination management. In terms of network performance, compared with Transyt and Synchro, the signal control scheme obtained by the proposed model can reduce the average delay by about 30% and the average number of stops by about 40%. To meet the requirements of regional green wave coordination control, the proposed model can optimize different signal timing parameters to obtain a wide range of applications of green wave coordinated control in road networks.

Overlapping phase and split phase are two typical phase patterns at signalized intersections in practice. The existing bandwidth-oriented progression control schemes mainly focus on overlapping phase to adopt the various layout of intersections and volumes rather than split phase. To address this issue, a two-way signal progression model, namely Pband, is developed to simultaneously optimize phase pattern choice from overlapping phase or split phase, phase sequence, and offsets. Also, the proposed Pband is formulated with a mixed integer nonlinear programming technique to guarantee an optimal solution. A numerical test based on a field arterial from the county of Jiashan, Jiaxing, China is employed to validate Pband under designed traffic patterns and signal timings. Numerical results have indicated that the Pband can show promise in increasing progression bandwidths and reducing average travel delay and the number of stops via traffic simulation experiments compared with the conventional Multiband model.

Traffic congestion has become a major problem restricting the development of major cities. ITS (Intelligent Transportation System) can record the state of traffic and predict the future traffic state, then reasonably optimize the travel scheme, so as to achieve the purpose of alleviating traffic congestion. Meanwhile, traffic flow prediction can provide data support for ITS, so many researchers have done a lot of research on traffic flow prediction. Many researchers take the traffic network as an undirected graph, and use the GCN (Graph Convolution Network) model to study the traffic flow prediction, and have achieved good prediction results. However, the traffic network is directed, and the traffic network is regarded as an undirected graph, which loses the direction information of the road network. Therefore, this inspires us to propose a graph convolution operator DGCN (Directed GCN), which can make full use of the in degree and out degree information of each station in the traffic network. The experimental results show that the graph convolution neural network based on this operator has better prediction accuracy than the state-of-the-art models.