Development and validation of machine learning models for predicting functional outcome after low-dose alteplase in the extended time window for acute ischemic stroke

BackgroundThis study aims to develop machine learning (ML) models to predict 90-day functional outcomes for acute ischemic stroke (AIS) patients receiving thrombolysis with low-dose alteplase at 0.6 mg/kg between 4.5 and 9 h after symptom onset.MethodsWe conducted a retrospective analysis of AIS patients receiving thrombolysis between August 1, 2019 and August 31, 2023. Eligible patients were randomly divided into training and validation sets in a 7:3 ratio. Good functional prognosis at 90 days were defined as modified Rankin scale score (mRS) ≤2. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to select optimal features. Five ML algorithms were employed to construct prediction models. Model performance was evaluated using receiver operating characteristic (ROC) curves, area under the curve (AUC) value, decision curve analysis (DCA), and calibration curves. SHapley Additive exPlanations (SHAP) plot was applied to interpret the model predictions.ResultsA total of 202 patients were randomly divided into training (n = 142) and validation (n = 60) sets. The rate of poor functional prognosis at 90 days was 56.34% in the training set and 56.67% in the validation set. Random Forest (RF) model showed the best discriminative ability with the highest AUC of 0.854 in the validation set. Key predictive features included age, baseline systolic blood pressure, white blood cell count, baseline National Institutes of Health Stroke Scale (NIHSS) score, wake-up stroke, the absolute difference volume between the ischemic infarct and the penumbra, intracranial hemorrhage, hemorrhagic transformation classification, and occurrence of pneumonia.ConclusionThe RF-based ML model demonstrated clinical utility for post-intravenous thrombolysis risk stratification by identifying patients at higher risk of poor functional outcomes.

A multichannel MEG time–frequency analysis framework for detecting stage -specific effects of spatial distraction in visual-spatial working memory

IntroductionSpatial distraction can disrupt visual-spatial working memory (VSWM), but its stage-dependent effects on multichannel neural dynamics remain insufficiently characterized. This study presents a multichannel magnetoencephalography (MEG) time—frequency analysis framework to detect stage-specific oscillatory responses to spatial distraction during a VSWM task.MethodsMEG signals were recorded from healthy participants under Distractor and No-distractor conditions and analyzed across encoding, maintenance, and retrieval/decision epochs. Time–frequency power was estimated in the delta, theta, alpha, beta, and gamma bands, and condition differences were evaluated using sensor-level spatiotemporal cluster-based permutation testing and Bonferroni correction within each predefined epoch.ResultsThe proposed analysis revealed a clear stage-specific pattern, with the most prominent modulation occurring during maintenance. Specifically, distraction induced robust and sustained increases in theta-, alpha-, and beta-band power during the retention interval (all cluster-level p < 0.01). Theta activity increased rapidly after maintenance onset and remained elevated throughout the full maintenance period over bilateral temporal, and widespread parieto-occipital sensors, while alpha and beta enhancements also showed temporally continuous and spatially stable patterns across widespread sensor networks.DiscussionThese findings highlight sustained large-scale oscillatory modulation as a key neural signature of distraction during mnemonic maintenance. The study provides an interpretable multichannel signal-analysis perspective on distraction effects in working memory and offers a practical framework for stage-resolved analysis of brain dynamics in cognitive tasks.

Computer-based tree drawing test in adolescents and adults with depression

ObjectiveTo evaluate the value of the computer-based Tree Drawing Test in the auxiliary diagnosis of depressive disorders and to analyze the differences in the performance of adolescent and adult depression patients in the Tree Drawing Projection Test.MethodsThis study was conducted at Guo Yang County People’s Hospital in Anhui, China, and involved a total of 184 participants: 43 adults with depression, 82 adolescents with depression, and 59 healthy controls. The Tree Drawing Test and scale assessments were administered to patients with depressive disorders (adult group and adolescent group) and a control group. Computer image recognition and calculation techniques were used to analyze the results statistically.ResultsSignificant differences were observed between the adult depression group and the control group in terms of crown area, trunk area, total area, and HDRS scores (p < 0.001). Statistically significant differences were also found between the adult depression group and the adolescent depression group in terms of trunk area (p < 0.01), total area (p < 0.001), HDRS scores (p < 0.001), and HAMA scores (p < 0.01). The crown area (r = -0.261, p < 0.001), trunk area (r = -0.154, p = 0.037), total area (r = -0.285, p < 0.001), and HDRS scores in the Tree Drawing Test were significantly correlated.ConclusionThe computer-based Tree Drawing Test has certain value in the auxiliary diagnosis of depression. Future research should include larger sample sizes and participants from different regions and cultural backgrounds to further validate the generalizability and cultural adaptability of the Tree Drawing Test for depression assessment.

ADOPT model combined with structured health education alleviates the preoperative anxiety of patients undergoing preventive ileostomy

ObjectiveThis study aimed to evaluate the efficacy of the ADOPT (Attitude-Definition-Openmind-Plan-Try it out) model combined with structured health education in alleviating preoperative anxiety in patients undergoing preventive ileostomy for rectal cancer.MethodsThis is a randomized controlled trial. A total of 60 patients scheduled for temporary ileostomy were randomly assigned to either the control group (routine care) or the research group (ADOPT model combined with science popularization interventions). The research group received structured education via a multimedia resource library, including preoperative, intraoperative, and postoperative care guidance, alongside interactive support from a specialized healthcare team. Anxiety levels were assessed with the State-Trait Anxiety Inventory (STAI) at admission and preoperatively.ResultsAt baseline, no significant differences were observed in gender (P = 0.202), age (P = 0.052), or BMI (P = 0.798) between the two groups. Both groups exhibited comparable anxiety levels at admission. However, one hour before surgery, the research group showed significantly lower state anxiety (S-AI) scores and total anxiety scores compared to the control group (20 ± 0.48 vs 23 ± 0.37, p<0.001), while trait anxiety (T-AI) scores remained similar (p<0.05).ConclusionThe integration of the ADOPT model with structured health education effectively reduces preoperative anxiety in ileostomy patients, highlighting its potential as a standardized nursing intervention.

Development and validation of a comprehensive prevention-focused intervention package for problematic digital technology use among youth: a multi-site study protocol

BackgroundProblematic use of digital technology among children, adolescents, and young adults is associated with adverse health, behavioural, interpersonal, social, academic and vocational outcomes. Most existing research focuses on treatment oriented interventions. Prevention focused interventions are limited. This is especially true for the low- and middle-income countries. There is a need for structured prevention approaches that involve youth, parents, and teachers.ObjectivesThis study aims to develop and validate a comprehensive package of prevention-focused interventions targeted at problematic use of digital technology among youth.MethodsThe study will be conducted across six sites in India. It will use a sequential mixed-methods design. Literature review, stakeholder interviews, and expert consensus shall be used for intervention development. This will be guided by established frameworks for complex interventions. Validation will be carried out using a quasi-experimental pre–post design. Quantitative measures will assess changes in knowledge, skills, confidence, and decision-making, as well as feasibility and acceptability. Qualitative methods will be used to assess engagement, delivery quality, and contextual factors.Expected outcomesThe study will lead to a modular prevention-focused intervention package with evidence of feasibility and acceptability. Findings will inform future larger scale implementation and evaluations.ConclusionThis protocol outlines a structured approach to development of a prevention-focused intervention targeted at problematic digital technology use among youth. The focus on prevention, stakeholder involvement, and real-world settings supports relevance for public health practice and policy.Clinical trial registrationhttps://ctri.nic.in/Clinicaltrials/login.php, identifier CTRI2026/03/105278.

Adaptation of behavioural activation for adolescents with mild to moderate intellectual disabilities and depression

IntroductionAdolescents with intellectual disabilities are at increased risk for mental health problems and depression. Despite this, there is currently no evidence for effective psychological interventions for treating low mood and depression in this population. Behavioural activation has been identified as an effective intervention for treating depression in autistic adolescents and for adults with intellectual disabilities and may therefore also be suitable for use with adolescents with intellectual disabilities.MethodThe current paper describes an approach taken to adapting an existing behavioural activation intervention used with adults with intellectual disabilities (Beat-It) to be suitable for adolescents, named Beat-Depression (Beat-D). An iterative, three-phase approach was adopted for the adaptation process. The first phase involved review of the Beat-It manual and proposed adaptations by the project team, followed by a second phase consisting of consultations with parents of adolescents with intellectual disabilities and professionals with experience in the field.ResultsThe outcomes from phases one and two were incorporated into a final adapted manual for the Beat-D intervention. The intervention is described following the principles of the Template for Intervention Description and Replication (TIDieR) checklist.DiscussionImplications for using this adaptation approach more broadly to ensure psychological interventions used with adolescents with intellectual disabilities are suitable and accessible are discussed along with future plans for the evaluation of Beat-D.

Virtual reality-based inhibition training influences food-related responses: no additional effects of repetitive transcranial magnetic stimulation

Combining cognitive inhibition training with brain stimulation techniques has received increasing attention as a potential approach to modulating maladaptive food craving and food intake. Building on previous work in this line of research, the current study examined whether virtual reality (VR)-based no-go inhibition training paired with repetitive transcranial magnetic stimulation (rTMS) modulates implicit food-related attitudes, craving and food-choice behaviors. Healthy women with high trait food cravings and a preference for high-calorie foods were assigned to one of four groups in a 2 (rTMS: active vs. sham) × 2 (training: no-go vs. neutral) between-subjects design. High-frequency rTMS was applied over the left dorsolateral prefrontal cortex (DLPFC), and no-go training was implemented in a VR environment using food stimuli tailored to participants’ self-reported preferences. Implicit attitudes and food craving were assessed before and after the intervention, while food choice was measured post-intervention only. Following training, the no-go group showed reduced positive implicit attitudes toward high-calorie foods and increased craving for low-calorie foods compared to pre-training levels, whereas no such changes were observed in the neutral group. Moreover, compared to the neutral group, the no-go group made healthier food choices. No-go training effects on food choice were more pronounced among individuals with low-to-moderate baseline preferences for high-calorie foods. In contrast, no significant main effects or additive effects of rTMS were observed. The present study demonstrates that VR-based no-go training can effectively regulate food-related responses and extends earlier work by demonstrating robust inhibition training effects across implicit and explicit measures, while highlighting the importance of considering individual differences in future research.

Here’s how technology transformed babymaking

Technology is changing the way we make babies. The pioneering work of the scientists who invented IVF led to the birth of the first “test tube baby” in 1978. We’ve come a long, long way since then.

This week, I’ve been working on a piece about the cutting edge of IVF technologies and what’s coming next. Think AI and robots and, potentially, gene-edited embryos.

My reporting has also made me think about just how much progress has been made in the last five decades. Clinicians have improved hormonal treatments. Embryologists have devised ways to culture embryos in the lab for longer. IVF clinics today offer multiple genetic tests for embryos.

In recent years, we’ve had reports of babies born with DNA from three people, babies born following “IVF on wheels,” babies born from decades-old embryos, and even babies “conceived” with the aid of a sperm-injecting robot.

The technology has also had a huge social impact. It has allowed for changes in the structure of families and provided more reproductive choices for would-be parents. So this week, let’s consider the technologies that have transformed babymaking.

Alan Penzias, a reproductive endocrinologist at Boston IVF, has been working in IVF since the early 1990s. In those days, his lab at Yale would collect a person’s eggs, fertilize them, and culture any resulting embryos for two days, until the embryos had two or four cells.

The embryos couldn’t survive any longer outside a body, so they’d be transferred to the uterus at that point. All of them. Even if there were, say, five embryos in total. Typical healthy patients could expect a live birth rate of 12% to 15%, he says.

Then Penzias heard that other teams were managing to culture embryos for three days. “We thought, No, that’s not possible,” he recalls. He learned that scientists had achieved this by tinkering with the culture medium—the nutrient-rich fluid the embryos are grown in.

Those three-day embryos, which had around six to 10 cells, seemed to have a better chance of resulting in a live birth. The teams culturing embryos for longer saw their success rates climb to 25% among similar patient groups, says Penzias. Again, he couldn’t believe it. “We thought they were making it up,” he says.

In the years since, teams have made more improvements to culture medium. Today, most IVF embryos are cultured for five or six days—a point at which they have 80 to 100 cells. The culturing process can act a little like a stress test—the embryos that make it to day six are generally more likely to go all the way and develop into a healthy baby.

Over the same period, advances in other technologies have opened up the options for what we can do with those embryos. Scientists learned they were able to freeze embryos and use them at a later date. A little over a decade ago, clinics shifted to a “vitrification” approach that rapidly cools the embryos to a glassy state. Vitrified embryos are more likely to survive freezing and thawing, so this approach quickly caught on.

As a result, doctors no longer needed to transfer multiple embryos at once. This made it less likely that patients would have twins or triplets, which can increase the risk of pregnancy complications.

Vitrification has also made IVF safer in other ways, including by affording patients a bit of time between fertility treatments. The hormonal treatments used in the first phase of IVF are designed to increase the production of mature eggs that can be collected. These treatments carry a small risk of a condition called ovarian hyperstimulation syndrome (OHSS), which in rare cases can be life-threatening. The ability to freeze all your embryos and use them at a later date is thought to give the body a chance to recover from hormonal treatment and reduces the risk of OHSS.

And because clinics are now able to culture embryos for up to a week, they can take a few of the 100 or so cells and send them for genetic testing before freezing the embryos. People undergoing IVF can get genetic readouts of all the embryos before deciding which to implant. (It is worth noting, however, that these testing technologies are not perfect.)

“Those are really radical changes, and we take them for granted,” says Penzias.

These technologies have also changed the function of IVF. What was once a treatment for infertility is now used to preserve fertility. People who want to delay parenthood can opt to freeze their eggs or embryos and use them later. They might opt to transfer one embryo in a year’s time and a second several years later. “We’ve been able to empower women to be able to have much more reproductive choice and get more reproductive mileage from a single IVF cycle,” says Penzias.

People who are about to undergo cancer treatments that might damage the testes or ovaries can opt to store their eggs or sperm ahead of time, too. Scientists have even been able to preserve pieces of ovarian and testicular tissue and reimplant them later, enabling recipients to have healthy babies.

Today, more people than ever have access to safe IVF options that offer multiple paths to parenthood. Those options look set to expand. But if you want to find out more about the AI and IVF robots, you’ll have to read this week’s story, here!

This article first appeared in The Checkup, MIT Technology Review’s weekly biotech newsletter. To receive it in your inbox every Thursday, and read articles like this first, sign up here.