Due to limitation of labeled resources, crossdomain named entity recognition (NER) has been a challenging task. Most existing work considers a supervised setting, making use of labeled data for both the source and target domains. A disadvantage of such methods is that they cannot train for domains without NER data. To address this issue, we consider using cross-domain LM as a bridge cross-domains for NER domain adaptation, performing crossdomain and cross-task knowledge transfer by designing a novel parameter generation network. Results show that our method can effectively extract domain differences from crossdomain LM contrast, allowing unsupervised domain adaptation while also giving state-ofthe-art results among supervised domain adaptation methods.

The techniques of proteomic analysis combined with tissue microarray provide us a dramatic tool for screening of LNM-associated proteins in cancer research. The increased expression of HSP-27, GST, and Annexin II, but decreased expression of L-FABP, suggests a significantly elevated incidence of LNM in CRC.

AimThe aim of this study was to examine the acceptability and efficacy of interactive short message service (SMS) in improving medication adherence in antiretroviral treatment (ART)-naïve individuals living with HIV/AIDS in Hengyang, Hunan, China.BackgroundSMS via mobile phone has emerged as a potential tool for improving ART adherence. However, most studies used SMS only as a medication reminder, with few studies exploring the effect of comprehensive, interactive SMS.Patients and methodsIn a randomized controlled trial, 100 HIV-positive patients on ART for <3 months were randomized into control or intervention arm. Participants in the control group received routine standard instruction for ART medication in the HIV clinics, while the intervention group received 6 months of an SMS intervention in addition to the standard care. A total of 124 text messages within 6 modules were edited, preinstalled, and sent to participants according to personalized schedules. Knowledge (of HIV and HIV medications), self-reported antiretroviral adherence (Visual Analog Scale [VAS] and Community Programs for Clinical Research on AIDS [CPCRA] Antiretroviral Medication Self-Report), and CD4 count were assessed at baseline and immediate post-intervention. Intervention participants were interviewed after completion of the study about their satisfaction with and acceptability of the SMS intervention.ResultsBaseline assessments were comparable between arms. Repeated-measures analysis showed that both HIV-related and ART medication knowledge of the intervention group showed better improvement over time than those of the control group after the intervention (P<0.0001). For the adherence measures, compared with the control group, participants in the intervention group had significantly higher VAS mean score (Z=2.735, P=0.006) and lower suboptimal adherence rate (Z=2.208, P=0.027) at the end of the study. The intervention had no effect on CD4 cell count. Almost all (96%) intervention participants reported satisfaction or high satisfaction with the SMS intervention, with 74% desiring to continue to receive the SMS intervention. The preferred frequency of messages was 1–2 messages per week.ConclusionAn interactive SMS intervention with comprehensive content shows promising efficacy in promoting medication adherence in ART-naïve individuals. Future work might further refine its ability to optimally tailor the intervention for individual preferences.