ChatGPT has sparked extensive discussions within the healthcare community since its November 2022 release. However, potential applications in the field of psychiatry have received limited attention. Deep learning has proven beneficial to psychiatry, and GPT is a powerful deep learning‐based language model with immense potential for this field. Despite the convenience of ChatGPT, this advanced chatbot currently has limited practical applications in psychiatry. It may be used to support psychiatrists in routine tasks such as completing medical records, facilitating communications between clinicians and with patients, polishing academic writings and presentations, and programming and performing analyses for research. The current training and application of ChatGPT require using appropriate prompts to maximize appropriate outputs and minimize deleterious inaccuracies and phantom errors. Moreover, future GPT advances that incorporate empathy, emotion recognition, personality assessment, and detection of mental health warning signs are essential for its effective integration into psychiatric care. In the near future, developing a fully‐automated psychotherapy system trained for expert communication (such as psychotherapy verbatim) is conceivable by building on foundational GPT technology. This dream system should integrate practical ‘real world’ inputs and friendly AI user and patient interfaces via clinically validated algorithms, voice comprehension/generation modules, and emotion discrimination algorithms based on facial expressions and physiological inputs from wearable devices. In addition to the technology challenges, we believe it is critical to establish generally accepted ethical standards for applying ChatGPT‐related tools in all mental healthcare environments, including telemedicine and academic/training settings.

Objectives To compare cognitive effects and acceptability of repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) in patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI), and to determine whether cognitive training (CT) during rTMS or tDCS provides additional benefits. Methods Electronic search of PubMed, Medline, Embase, the Cochrane Library and PsycINFO up to 5 March 2020. We enrolled double-blind, randomised controlled trials (RCTs). The primary outcomes were acceptability and pre–post treatment changes in general cognition measured by Mini-Mental State Examination, and the secondary outcomes were memory function, verbal fluency, working memory and executive function. Durability of cognitive benefits (1, 2 and ≥3 months) after brain stimulation was examined. Results We included 27 RCTs (n=1070), and the treatment components included high-frequency rTMS (HFrTMS) and low-frequency rTMS, anodal tDCS (atDCS) and cathodal tDCS (ctDCS), CT, sham CT and sham brain stimulation. Risk of bias of evidence in each domain was low (range: 0%–11.1%). HFrTMS (1.08, 9, 0.35–1.80) and atDCS (0.56, 0.03–1.09) had short-term positive effects on general cognition. CT might be associated with negative effects on general cognition (−0.79, –2.06 to 0.48) during rTMS or tDCS. At 1-month follow-up, HFrTMS (1.65, 0.77–2.54) and ctDCS (2.57, 0.20–4.95) exhibited larger therapeutic responses. Separate analysis of populations with pure AD and MCI revealed positive effects only in individuals with AD. rTMS and tDCS were well tolerated. Conclusions HFrTMS is more effective than atDCS for improving global cognition, and patients with AD may have better responses to rTMS and tDCS than MCI.

Importance Although several pharmacological interventions for delirium have been investigated, their overall benefit and safety remain unclear. Objective To evaluate evidence regarding pharmacological interventions for delirium treatment and prevention. Data Sources PubMed, Embase, ProQuest, ScienceDirect, Cochrane Central, Web of Science, ClinicalKey, and ClinicalTrials.gov from inception to May 17, 2018. Study Selection Randomized clinical trials (RCTs) examining pharmacological interventions for delirium treatment and prevention. Data Extraction and Synthesis To extract data according to a predetermined list of interests, the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines were applied, and all meta-analytic procedures were conducted using a random-effects model. Main Outcomes and Measures The primary outcomes were treatment response in patients with delirium and the incidence of delirium in patients at risk of delirium. Results A total of 58 RCTs were included, in which 20 RCTs with 1435 participants (mean age, 63.5 years; 65.1% male) compared the outcomes of treatment and 38 RCTs with 8168 participants (mean age, 70.2 years; 53.4% male) examined the prevention of delirium. A network meta-analysis demonstrated that haloperidol plus lorazepam provided the best response rate for delirium treatment (odds ratio [OR], 28.13; 95% CI, 2.38-333.08) compared with placebo/control. For delirium prevention, the ramelteon, olanzapine, risperidone, and dexmedetomidine hydrochloride groups had significantly lower delirium occurrence rates than placebo/control (OR, 0.07; 95% CI, 0.01-0.66 for ramelteon; OR, 0.25; 95% CI, 0.09-0.69 for olanzapine; OR, 0.27; 95% CI, 0.07-0.99 for risperidone; and OR, 0.50; 95% CI, 0.31-0.80 for dexmedetomidine hydrochloride). None of the pharmacological treatments were significantly associated with a higher risk of all-cause mortality compared with placebo/control. Conclusions and Relevance This network meta-analysis demonstrated that haloperidol plus lorazepam might be the best treatment and ramelteon the best preventive medicine for delirium. None of the pharmacological interventions for treatment or prophylaxis increased the all-cause mortality.