Because of their unique adaptability, flexibility, and robustness, musculoskeletal robotic systems are regarded potentially as next-generation robots. However, motion learning and generation of such a robotic system are still challenging. This paper presents a neuromuscular control method, namely, TMS-PPO, based on time-varying muscle synergy (TMS) and proximal policy optimization (PPO). The electromyogram (EMG) activation signals of actual human motions are decomposed to obtain TMSs based on the temporal properties of the TMS. The weights of networks are trained to generate the scale and phase coefficients through the PPO. The coefficients modulate the TMSs to generate appropriate activation patterns to optimize motion learning of the musculoskeletal system. To verify the effectiveness of the proposed method, the TMSs are extracted from human upper limb muscle activation signals, and we compare TMS-PPO with PPO in the motion learning and generation process of an upper limb musculoskeletal system. The results show that TMS-PPO can complete the control tasks because the average errors of the joints are less than 0.05 rad. In the meantime, TMSs are used as motion primitives of the musculoskeletal system to simulate the process of the human CNS controlling muscles. It shows that TMS-PPO reduces the energy consumption and improves the learning rate significantly compared with the PPO. The learning episodes reduce from \(10^4\) to \(10^3\) , which indicates that TMS-PPO has a stronger learning ability and better physiological explanation. Note to Practitioners —Due to the superiorities of the musculoskeletal system, humanoid robots that imitate human driven mechanisms are vigorously carried out worldwide. Taking advantages of human-like characteristics, the musculoskeletal robot provides new opportunities to understand and validate the human mechanisms of muscle control and motion learning, to compare the performance of the robot to that of humans as well as work in real world, e.g., human interactive robots, amusement robots and medical training robots in the future. However, strong redundancy, coupling, and nonlinearity of the system also raises many challenges for the investigation of the control problem. Inspired by how the human CNS controls a musculoskeletal system and realize motion generalization, a novel muscle-synergies-based neuromuscular control that combines time-varying muscle synergy (TMS) and Proximal Policy Optimization (PPO), namely, TMS-PPO is proposed in this paper. The learning efficiency of PPO and the physiological interpretation of the control process are improved during the motion learning and generation processes of the musculoskeletal system. Preliminary simulation experiments suggest that this method is feasible in terms of control accuracy and efficiency. Moreover, the performance of the TMS-PPO is comparable to the PPO without significant improvement. To solve this problem, in future work, we will introduce the cerebellar model into the control method which plays the role of adjusting and correcting the motions of the limbs to achieve accurate and stable control in the actions process of humans.

Depression represents one of the most prevalent mental health disorders globally, significantly impacting quality of life and posing substantial healthcare challenges. Traditional diagnostic methods rely on subjective assessments and clinical interviews, often leading to misdiagnosis, delayed treatment, and suboptimal outcomes. Recent advances in biosensing technologies offer promising avenues for objective depression assessment through detection of relevant biomarkers and physiological parameters. This review examines multi-modal biosensing approaches for depression by analyzing electrochemical biosensors for neurotransmitter monitoring alongside wearable sensors tracking autonomic, neural, and behavioral parameters. We explore sensor fusion methodologies, temporal dynamics analysis, and context-aware frameworks that enhance monitoring accuracy through complementary data streams. The review discusses clinical validation across diagnostic, screening, and treatment applications, identifying performance metrics, implementation challenges, and ethical considerations. We outline technical barriers, user acceptance factors, and data privacy concerns while presenting a development roadmap for personalized, continuous monitoring solutions. This integrative approach holds significant potential to revolutionize depression care by enabling earlier detection, precise diagnosis, tailored treatment, and sensitive monitoring guided by objective biosignatures. Successful implementation requires interdisciplinary collaboration among engineers, clinicians, data scientists, and end-users to balance technical sophistication with practical usability across diverse healthcare contexts.

SUMMARY: In nature, sounds from multiple sources, as well as reflections from the surfaces of the physical surroundings, arrive concurrently from different directions at the ears of a listener. Despite the fact that all of these waveforms sum at the eardrums, humans with normal hearing can effortlessly segregate interesting sounds from echoes and other sources of background noises. This paper presents a two-microphone technique for localization of sound sources to effectively guide robotic navigation. Its fundamental structure is adopted from a binaural signal-processing scheme employed in biological systems for the localization of sources using interaural time differences (ITDs). The two input signals are analyzed for coincidences along left/right-channel delay-line pairs. The coincidence time instants are presented as a function of the interaural coherence (IC). Specifically, we build a sphere head model for the selected robot and apply the mechanism of binaural cues selection observed in mammalian hearing system to mitigate the effects of sound echoes. The sound source is found by determining the azimuth at which the maximum of probability density function (PDF) of ITD cues occurs. This eliminates the localization artifacts found during tests. The experimental results of a systematic evaluation demonstrate the superior performance of the proposed method.

While the sensors of current EEG wearables can be miniaturized such that they can be fit into an in-ear hearable device, the brain monitoring based on the state-of-art ear-EEG techniques has several drawbacks such as customization for one individual, time-consuming fabrication process and short life cycle. To deal with this problem, a new method to design a sensor with one channel dry electrode embedded in a generic substrate is proposed. The electrical characterizations were evaluated with 15 subjects and five EEG paradigms, in which ear-EEG and conventional scalp EEG was recorded simultaneously. Recordings from the generic dry-contact ear-EEG sensor were compared to conventional scalp EEG recordings. Reliable responses were extracted from the ear-EEG for different paradigms, albeit weaker in amplitude than from scalp EEG. The results demonstrate the feasibility of the designed generic ear-EEG sensor and its potential to achieve sustained, ease-of-use, cost-effective and off-the-shelf ear-EEG recording use.

Porous metals are found to be suitable as orthopedic scaffolds. However, only limited control over the internal architecture can be achieved using conventional design methods. The architecture with porosity variation strategy mimicking natural bone is critical to gain favorable combination of mechanical and biological properties for orthopedic implants. In this regard, a topology optimization method with customized morphology and mechanical properties derived from the trabecular bone was proposed to design three-dimensional architectures with gradient porosity resembling the porous structure of bone. In particular, the elastic constants for the trabecular bone were better predicted when the bone volume fraction was supplemented with a three-dimensional structural parameter, i.e., degree of anisotropy. These constants were set as the optimization constrains for morphology control. Then the porous titanium structures were manufactured by selective laser melting technology (SLM). The physical characteristics, mechanical properties of the scaffolds were compared systematically. The experimental results revealed that the as-built samples with the proposed method lead to a good match of morphological accuracy and mechanical properties to that of the bone. It demonstrates that the proposed topology optimization method with controlled morphology and mechanical properties provides an efficient manner for the biomimetic design of orthopedic implants.

The cerebellum plays an important role in smooth and coordinated motor control. Precise control requires the cerebellum to regulate movements in both space and time domains. Although many cerebellar models have been proposed, most of them focus on motor coordination, motor learning, or timing distinctly. Therefore, it is necessary to develop a cerebellar model which contributes to the proper execution of movements via motor learning that displays temporal specificity. In this paper, we proposed a novel spiking neural network model to realize cerebellar processing with strong biomimicry, which is based on adapting rate neurons and has a cerebellum-inspired structure as well as biologically plausible cerebellar divergence/convergence ratios. The model was tested with the eyeblink classical conditioning (EBCC) experimental task. The simulation results verify that our implementation has improvements in both signal encoding and learning speed than previous studies. Furthermore, compared with neurophysiological data, our model shows a similar learning trend and can represent the fluctuation of the learning curves. It is demonstrated that our proposed cerebellar model experimentally reproduces the key features of the EBCC task and provides a new way to understand the timing and learning control neural mechanisms of the cerebellum.