Here you can find a list of publishable project dissemination items:
2023
Low-intensity focused ultrasound is a promising non-invasive neuromodulation technique for depression that enables precise stimulation of deep brain networks. Unlike existing therapies, it offers high spatial resolution and broader brain coverage, but it is still in early development and requires further research to confirm clinical effectiveness
Depression arises from complex and highly variable brain network dysfunctions, which limits the effectiveness of current brain stimulation technologies. Because symptoms and underlying neural circuits differ widely between individuals, precise and personalized targeting remains challenging. Emerging approaches, such as advanced neuromodulation techniques, aim to better address this variability and improve treatment outcomes.
Low-intensity focused ultrasound (LIFU) is an emerging, non-invasive technique that can precisely target and modulate neural circuits involved in depression. By directly influencing brain networks rather than isolated regions, it offers potential for personalized treatment, but its clinical application remains in early stages and requires further validation.
Power and area efficient on-chip feature extraction is needed for future closed-loop neural interfaces. This paper presents a feature extraction unit for neural oscillatory synchrony that bypasses the phase extraction step to reduce hardware complexity. Instead, the sine and cosine of the phase are directly approximated from the real and imaginary components of the signal to calculate the phase-amplitude coupling (PAC) and phase locking value (PLV). The synthesized design achieves state-of- the-art performances at 43 nW/channel and 0.006 mm2, while maintaining sufficient accuracy for seizure detection in epileptic patients.
Published in: 2023 IEEE International Symposium on Circuits and Systems (ISCAS)
Responsive neuromodulation faces limitations due to mismatches between biological ion-based signaling and conventional electron-based materials, leading to inefficiencies at neural interfaces. Conducting polymers offer improved performance by enabling ion penetration and reducing impedance, while advances in CMOS-based electronics provide high-density, energy-efficient neural interfaces. Integrating these material and technological innovations is essential to enhance neurostimulation systems and bridge current gaps in device design.
Understanding and modulating neural networks requires high-resolution acquisition of neural activity over time, real-time analysis, and minimally invasive stimulation methods with high specificity. Such procedures are particularly needed for treatment of sensory disfunction (e.g. hearing loss), and certain neurological diseases (e.g. epilepsy). The lack of soft, biocompatible, hybrid and smart neural interfaces hinders our capacity to study complex neural dynamics and efficiently apply responsive neuromodulation therapy. Here, I am presenting the vision of Neural Waves lab towards designing and developing materials and novel fully implantable, contained and responsive neural interfaces that will allow long-term acquisition and closed-loop manipulation of neural circuits with high spatiotemporal resolution over extended period of time to reveal neural dynamics in different neurological pathways and alleviate disfunctions and diseases. I will cover our studies on i) creating artificial basilar membrane based on acousto-sensitive ion-based transistors and soft electronics, ii) innovating electroencephalography interfaces, and iii) utilizing organic perforated multielectrode arrays to investigate the effect of photopharmacological interventions on Cortical Seizures.
Published in: Materials for Sustainable Development Conference (MATSUS)
Proceedings of MATSUS Fall 2023 Conference (MATSUSFall23)
Brain signals show promise as biomarkers for depression by reflecting underlying neural network dysfunctions and enabling more objective diagnosis and monitoring. Although current findings highlight their potential to guide personalized treatments and improve neuromodulation strategies, challenges remain in achieving reliable, clinically applicable biomarkers.
2024
Piezoelectric Micromachined Ultrasonic Transducers (PMUT) and Capacitive Micromachined Ultrasonic Transducers (CMUT) have seen great developments in recent years, both in terms of performance and scope of applications within the biomedical ultrasound domain. This paper presents a review of the state-of-the-art of PMUT and CMUT technologies, focusing on their principle of operation, microfabrication techniques and use in different biomedical imaging and therapeutic applications. The advantages and drawbacks of PMUT and CMUT technologies in comparison with conventional bulk transducers are highlighted, the trade-offs among PMUTs and CMUTs are discussed, and their relevance in the current landscape of medical diagnostics and therapeutic uses is outlined, thus providing a clear overview of these promising technologies for the present and the next generation biomedical ultrasound applications.
Published in: IEEE Access (Volume: 12)
Epilepsy is a chronic neurological disorder marked by recurrent seizures, with current treatments often limited by side effects and incomplete efficacy. Targeting the adenosine type 1 receptor presents a promising approach for seizure suppression, but it may also lead to adverse effects such as dizziness, sleep disturbances, and cognitive or mood changes.
Bulk piezoelectric ultrasound transducers on integrated circuits offer unique properties for therapeutic applications of ultrasound neuromodulation. However, current implementations of such transducers are not optimized for the high transmit efficiency required to stimulate neurons. This is mainly due to the challenge of implementing a metal layer on top of the piezoelectric film using microfabrication techniques. Here, we propose a micromachined capping structure providing an electrical connection on top of the piezoelectric film with minimal acoustic losses. The structure can potentially be used as a common ground connection in phased-array ultrasound transducers.
Published in: Proceedings of XXXV EUROSENSORS Conference
Reserpine (RES), a Vesicular Monoamine Transporter 2 (VMAT2) inhibitor agent, has been used in preclinical research for many years to create animal models for depression and to test experimental antidepressant strategies. Nevertheless, evidence of the potential use and validity of RES as a chronic pharmacological model for depression is lacking, and there are no comprehensive studies of the behavioral effects in conjunction with molecular outcomes.
Published in: Progress in Neuro-Psychopharmacology and Biological Psychiatry Volume 133
Developing an implantable/wearable 2D ultrasound phased array for ultrasound neuromodulation poses several challenges, including power requirements for driving the piezoelectric transducers to generate sufficient pressure at the focal spot. Therefore, minimizing power consumption is crucial to minimize excessive thermal dissipation and to ensure long-term usability without frequent charging or battery replacement. Prior work has improved efficiency based on transducer fabrication and circuit design optimizations. To further address this issue, we propose a new approach to minimize power consumption by tailoring the driving amplitude of each element in a 2D phased array based on their individual contribution to the focal spot pressure.
Published in: 2024 IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium (UFFC-JS)
Medial Forebrain Bundle Deep Brain Stimulation (MFB-DBS) can have rapid and long lasting antidepressant effects in Treatment Resistant Depression (TRD) patients. The mechanisms are not well understood, but one hypothesis stipulates that modulation of the dopaminergic (DAergic) fibers contribute to the therapeutic outcome. Acute DBS effects on DA release have been studied; however, longitudinal studies with acute-repetitive DBS are lacking. Long-Evans accumbal DA release and Ventral Tegmental Area (VTA) calcium tonic and phasic signaling to different mfb-DBS parameters were measured using fiber photometry over 8 weeks, following acute and repetitive stimulation in behaving and non-behaving animals. DBS-induced release was observed in both targets, with increased frequency and DBS duration. 130 Hz stimulation increased phasic and tonic DA response over time, with the latter being a potential mechanism for its long-term clinical effectiveness. VTA calcium transients decreased, while phasic activity increased with frequency. Pulse width (PW)-mediated differential peak release timing also suggests potential parallel activation of diverse fiber types. Additionally, decreased DA transients rate during Elevated Plus Maze (EPM) suggests context and stimulation duration-dependent DA release. The data confirm chronic antidromic/orthodromic DAergic responses with stimulation parameter dependent variability, providing novel insights into temporal adaptations, connectivity and fiber recruitment on mfb DBS.
Published in: Journal of Neurochemistry
2025
Medial forebrain bundle deep brain stimulation (mfb-DBS) produces antidepressant effects partly through modulation of GABAergic signaling rather than direct dopaminergic activation. In contrast to selective optogenetic stimulation, mfb-DBS increased GABA-related biomarkers in frontal and accumbal regions, with effects observed bilaterally even under unilateral stimulation. These changes were transient and not enhanced by repeated stimulation, suggesting that therapeutic outcomes may involve short-term network modulation influencing dopamine release.
Published in: Scientific Reports (Nature Portfolio)
Piezoelectric internal ion-gated organic electrochemical transistors (Piezo‑IGTs) enable efficient detection of physiological vibrations by directly converting mechanical signals into amplified electrical outputs. By integrating piezoelectric films with ion-sensitive transistors, these devices achieve high signal-to-noise ratios, low-power operation, and compact form factors. Their flexible, self-contained design supports accurate bio-mechanical monitoring in applications such as speech recognition and cardiovascular sensing, highlighting strong potential for wearable, implantable, and neuroprosthetic systems.
Published in: npj Flexible Electronics (Nature Portfolio)
This work presents a closed-loop neuromodulation chipset, integrating 64 analog frontends (AFEs) featuring a novel IDAC-embedded OTA, a lightweight feature extraction unit, a Sparse Projection Oblique Randomer Forest (SPORF) classifier, and 4 high-voltage (HV) compliant current stimulators. Each AFE occupies 0.0009mm2 and consumes 0.36μW, the smallest area and lowest power reported to date for ECoG recording (BW≤1kHz). Verified with CHB-MIT [1] and ETHZ [2] databases, the digital backend (DBE) occupies 0.075mm2, consumes 6.76μW and achieves event-based sensitivity and specificity of 94.93%/98.58% and 99.55%/99.97%, respectively, in patient-specific scenarios.
Published in: IEEE conference proceedings (via IEEE Xplore)
Medial forebrain bundle deep brain stimulation (mfb‑DBS) exerts antidepressant effects partly through modulation of central noradrenergic (NA) signaling and GABAergic inhibitory circuits. In a rodent depression model, DBS enhanced noradrenaline activity in key regions and normalized disrupted interneuron function, particularly in the prefrontal cortex and nucleus accumbens. These changes, alongside behavioral improvements, suggest that combined regulation of NA transmission and neural excitability contributes to the therapeutic mechanisms of DBS in depression.
Published in: Translational Psychiatry (Nature Portfolio)
This work presents a novel ultrasound system architecture that enables both imaging and therapy using a single transducer array by electronically controlling the Q‑factor. Through active damping compensation, the system dynamically reduces Q‑factor for imaging while maintaining a high-Q state for therapy, eliminating the need for separate transducers. Simulations demonstrate effective and stable Q-factor tuning with minimal variation, enabling compact, efficient, and versatile dual-mode ultrasound systems.
Published in: IEEE conference proceedings (via IEEE Xplore)
Emerging biomedical ultrasound applications such as pulsed neurostimulation and shear-wave imaging demand single-pulse focused ultrasound waves with MPa-range acoustic pressures. Achieving high pressures typically involves driving transducers with high voltages, necessitating bulky power amplifiers. Recently, phased arrays have emerged to miniaturize these focused transducers. However, they often exhibit poor power efficiency and heat dissipation. To address this, we explore acoustic amplification through acoustic energy storage and release, where, with minimal voltage, high-amplitude ultrasound waves are produced. Prior work has shown the principle using bulky apparatus with limited applicability. In this work, we explore the theory and perform finite element modeling (FEM) to investigate this mechanism with miniaturized and micro-electro-mechanical systems (MEMS)-compatible materials and geometries.
Published in: IEEE conference proceedings (via IEEE Xplore)
This study introduces a pseudo‑random pulse skipping driving scheme to improve power efficiency in ultrasound phased‑array systems for neuromodulation. By selectively reducing element-level output while maintaining stable focal pressure, the approach minimizes energy consumption without compromising therapeutic performance. Implemented in ASIC, the method enables controllable power modulation and avoids the drawbacks of existing techniques such as circuit complexity, switching losses, and pressure instability, making it suitable for implantable and wearable applications.
Published in: IEEE conference proceedings (via IEEE Xplore)
2026
This work presents a bidirectional neuromodulation chipset integrating low-noise neural recording and high-voltage stimulation within a heterogeneous CMOS architecture. By co-designing the analog front-end and digital backend, the system relaxes noise requirements while significantly reducing area and power consumption. A novel embedded current-steering DAC further improves signal fidelity and channel uniformity. The chipset achieves high-voltage stimulation capability alongside state-of-the-art efficiency in neural recording, demonstrating strong potential for compact, implantable closed-loop neuromodulation systems.
Published in: IEEE conference proceedings (via IEEE Xplore)
UPSIDE’s project poster as presented during the EIC Summit 2022
UPSIDE’s introductory leaflet from our team available for project’s dissemination purposes.
UPSIDE’s brand new Newsletter is out that summarizes our Year 1 progresses so far.
UPSIDE’s brand new Newsletter is out and summarizes our progress so far in Year 2.