The study presents a recursive least-squares estimation method with an exponential forgetting factor for noise removal in functional near-infrared spectroscopy data and extraction of hemodynamic responses (HRs) from the measured data. The HR is modeled as a linear regression form in which the expected HR, the first and second derivatives of the expected HR, a short-separation measurement data, three physiological noises, and the baseline drift are included as components in the regression vector. The proposed method is applied to left-motor-cortex experiments on the right thumb and little finger movements in five healthy male participants. The algorithm is evaluated with respect to its performance improvement in terms of contrast-to-noise ratio in comparison with Kalman filter, low-pass filtering, and independent component method. The experimental results show that the proposed model achieves reductions of 77% and 99% in terms of the number of channels exhibiting higher contrast-to-noise ratios in oxy-hemoglobin and deoxy-hemoglobin, respectively. The approach is robust in obtaining consistent HR data. The proposed method is applied for both offline and online noise removal.

Activated channels of functional near-infrared spectroscopy are typically identified using the desired hemodynamic response function (dHRF) generated by a trial period. However, this approach is not possible for an unknown trial period. In this paper, an innovative method not using the dHRF is proposed, which extracts fluctuating signals during the resting state using maximal overlap discrete wavelet transform, identifies low-frequency wavelets corresponding to physiological noise, trains them using long-short term memory networks, and predicts/subtracts them during the task session. The motivation for prediction is to maintain the phase information of physiological noise at the start time of a task, which is possible because the signal is extended from the resting state to the task session. This technique decomposes the resting state data into nine wavelets and uses the fifth to ninth wavelets for learning and prediction. In the eighth wavelet, the prediction error difference between the with and without dHRF from the 15-s prediction window appeared to be the largest. Considering the difficulty in removing physiological noise when the activation period is near the physiological noise, the proposed method can be an alternative solution when the conventional method is not applicable. In passive brain-computer interfaces, estimating the brain signal starting time is necessary.

Functional near-infrared spectroscopy (fNIRS) technology has been widely used to analyze biomechanics and diagnose brain activity. Despite being a promising tool for assessing the brain cortex status, this system is susceptible to disturbances and noise from electrical instrumentation and basal metabolism. In this study, an alternative filtering method, maximum likelihood generalized extended stochastic gradient (ML-GESG) estimation, is proposed to overcome the limitations of these disturbance factors. The proposed algorithm was designed to reduce multiple disturbances originating from heartbeats, breathing, shivering, and instrumental noises as multivariate parameters. To evaluate the effectiveness of the algorithm in filtering involuntary signals, a comparative analysis was conducted with a conventional filtering method, using hemodynamic responses to auditory stimuli and psycho-acoustic factors as quality indices. Using auditory sound stimuli consisting of 12 voice sources (six males and six females), the fNIRS test was configured with 18 channels and conducted on 10 volunteers. The psycho-acoustic factors of loudness and sharpness were used to evaluate physiological responses to the stimuli. Applying the proposed filtering method, the oxygenated hemoglobin concentration correlated better with the psychoacoustic analysis of each auditory stimulus than that of the conventional filtering method.

Neurovascular dysfunction is a hallmark of Alzheimer’s disease (AD), yet alterations in the temporal precision of the hemodynamic response remain difficult to detect using conventional trial-averaged approaches. Here we introduce a Single-Trial Kinetic Profiling (STKP) pipeline to analyze over 50,000 prefrontal hemodynamic events across healthy controls (HC), mild cognitive impairment (MCI), and AD. Single-trial latency (time-to-peak; Tmax) exhibited a graded rightward shift across the disease spectrum, with AD participants showing prolonged peak timing (mean ≈ 2.51 s vs. 1.63 s in HC) and increased trial-to-trial variability. A significant Group × Accuracy interaction indicated reduced task-dependent temporal modulation in the clinical groups. Whereas HC participants demonstrated modest latency adjustments during incorrect trials, AD responses extended to ≈ 2.88 s, suggesting diminished temporal flexibility under cognitive demand. Robustness analyses confirmed that these effects were preserved across alternative latency definitions. In leave-one-subject-out cross-validation, mean latency discriminated HC from AD with an AUC of 0.881 and HC from MCI with an AUC of 0.776, while amplitude-derived features did not improve performance. Together, these findings indicate that single-trial hemodynamic latency captures clinically relevant variance and may provide a sensitive marker of altered neurovascular timing along the Alzheimer’s continuum.

Background: Recent advances in digital therapeutics (DTx) have led to the development of game-based, home-delivered interventions for children with attention-deficit/hyperactivity disorder (ADHD). Prior studies have demonstrated the feasibility and clinical potential of DTx for reducing ADHD symptoms but have primarily evaluated usage metrics, including training frequency and duration, which do not capture in-game performance or learning processes during training. In addition, the association between task difficulty regulation during DTx training and ADHD symptom improvement in children has not been well studied. Objective: This study aimed to examine whether task difficulty regulation during DTx training is associated with ADHD symptom improvement in children and to evaluate the clinical relevance of difficulty-related behavioral measures as indicators of treatment response. Methods: In this study, a secondary post hoc analysis of data from 35 children aged 6 to 13 years with ADHD who participated in a randomized controlled trial was performed. The participants had completed a 4-week training program comprising 5 digital cognitive tasks from the DTx Neuro-World. During training, task difficulty was automatically adjusted in real time according to participants’ response accuracy to maintain an appropriate challenge level. Training difficulty was quantified by 3 indices: average task difficulty, task difficulty variability, and task difficulty slope. Associations between difficulty indicators and ADHD symptom changes were examined using regression analyses and group comparisons. Results: We found that lower average task difficulty during tasks 1, 2, and 4 of Neuro-World was associated with improvements in ADHD symptoms, as assessed by the total Korean-ADHD Rating Scale (K-ARS) score. In addition, the average task difficulty of tasks 1 to 4 was related to improvements in hyperactivity/impulsivity symptoms of ADHD, as assessed by the K-ARS hyperactivity or impulsivity subscore, but not to improvements in inattention symptoms. Moreover, lower variability of task difficulty in task 5 improved ADHD hyperactivity/impulsivity symptoms, as assessed by K-ARS, but did not affect Korean-Child Behavior Checklist scores. Finally, we found that a negative task difficulty slope was associated with the alleviation of ADHD hyperactivity or impulsivity. These findings should be interpreted with caution, given the exploratory nature of the study and the absence of formal correction for multiple comparisons. Conclusions: As an exploratory investigation, our study provides a novel perspective by demonstrating that individualized and stable regulation of task difficulty, rather than the mere attainment of higher difficulty levels, is associated with ADHD symptom improvement. These findings indicate that dynamic difficulty trajectories derived from in-game behavioral data are clinically relevant, going beyond the conventional focus in the literature on usage metrics or overall training intensity. Our results support the role of individualized difficulty trajectories as potential digital biomarkers of clinical response. This approach has important implications for the design of DTx that incorporate personalized and adaptive difficulty regulation to enhance clinical effectiveness and scalability.

The visual space is sampled by cortical field maps in which nearby neuronal populations encode nearby locations of images received from the retina. Whether retinotopic cortical organization already emerges in the neonatal or even prenatal human brain is currently unknown. To answer this question in vivo, we applied population connective field modeling to 871 resting-state functional magnetic resonance imaging datasets ranging from prenatal to young adult age. We found topographically organized eccentricity and polar angle connectivity maps in V2 and V3 of the visual cortex as early as the 21st week of gestation. These results highlight that human proto-retinotopic cortical maps develop in the second trimester of pregnancy, predating visual experience.