Research Highlights (Required)To create your highlights, please type the highlights against each \item command.It should be short collection of bullet points that convey the core findings of the article. It should include 3 to 5 bullet points (maximum 85 characters, including spaces, per bullet point.)• We survey deep learning based HAR in sensor modality, deep model, and application.• We comprehensively discuss the insights of deep learning models for HAR tasks.• We extensively investigate why deep learning can improve the performance of HAR.• We also summarize the public HAR datasets frequently used for research purpose.• We present some grand challenges and feasible solutions for deep learning based HAR. ABSTRACTSensor-based activity recognition seeks the profound high-level knowledge about human activities from multitudes of low-level sensor readings. Conventional pattern recognition approaches have made tremendous progress in the past years. However, those methods often heavily rely on heuristic hand-crafted feature extraction, which could hinder their generalization performance. Additionally, existing methods are undermined for unsupervised and incremental learning tasks. Recently, the recent advancement of deep learning makes it possible to perform automatic high-level feature extraction thus achieves promising performance in many areas. Since then, deep learning based methods have been widely adopted for the sensor-based activity recognition tasks. This paper surveys the recent advance of deep learning based sensor-based activity recognition. We summarize existing literature from three aspects: sensor modality, deep model, and application. We also present detailed insights on existing work and propose grand challenges for future research.

Despite success in hematologic malignancies, the treatment landscape of chimeric antigen receptor (CAR) T cell therapy for solid tumors remains limited. Claudin18.2 (CLDN18.2)-redirected CAR T cells showed promising efficacy against gastric cancer (GC) in a preclinical study. Here we report the interim analysis results of an ongoing, open-label, single-arm, phase 1 clinical trial of CLDN18.2-targeted CAR T cells (CT041) in patients with previously treated, CLDN18.2-positive digestive system cancers (NCT03874897). The primary objective was safety after CT041 infusion; secondary objectives included CT041 efficacy, pharmacokinetics and immunogenicity. We treated 37 patients with one of three CT041 doses: 2.5 × 108, 3.75 × 108 or 5.0 × 108 cells. All patients experienced a grade 3 or higher hematologic toxicity. Grade 1 or 2 cytokine release syndrome (CRS) occurred in 94.6% of patients. No grade 3 or higher CRS or neurotoxicities, treatment-related deaths or dose-limiting toxicities were reported. The overall response rate (ORR) and disease control rate (DCR) reached 48.6% and 73.0%, respectively. The 6-month duration of response rate was 44.8%. In patients with GC, the ORR and DCR reached 57.1% and 75.0%, respectively, and the 6-month overall survival rate was 81.2%. These initial results suggest that CT041 has promising efficacy with an acceptable safety profile in patients with heavily pretreated, CLDN18.2-positive digestive system cancers, particularly in those with GC.

No abstract

In activity recognition, it is often expensive and time-consuming to acquire sufficient activity labels. To solve this problem, transfer learning leverages the labeled samples from the source domain to annotate the target domain which has few or none labels. Existing approaches typically consider learning a global domain shift while ignoring the intra-affinity between classes, which will hinder the performance of the algorithms. In this paper, we propose a novel and general cross-domain learning framework that can exploit the intra-affinity of classes to perform intra-class knowledge transfer. The proposed framework, referred to as Stratified Transfer Learning (STL), can dramatically improve the classification accuracy for cross-domain activity recognition. Specifically, STL first obtains pseudo labels for the target domain via majority voting technique. Then, it performs intra-class knowledge transfer iteratively to transform both domains into the same subspaces. Finally, the labels of target domain are obtained via the second annotation. To evaluate the performance of STL, we conduct comprehensive experiments on three large public activity recognition datasets (i.e. OPPORTU-NITY, PAMAP2, and UCI DSADS), which demonstrates that STL significantly outperforms other state-of-the-art methods w.r.t. classification accuracy (improvement of 7.68%). Furthermore, we extensively investigate the performance of STL across different degrees of similarities and activity levels between domains. And we also discuss the potential of STL in other pervasive computing applications to provide empirical experience for future research.