This study bridges neuroscience and artificial intelligence by developing advanced models to predict cognitive states—specifically attention and meditation—using raw EEG data collected from low-cost commercial devices such as NeuroSky and Brainlink. Leveraging the temporal capabilities of recurrent neural networks (RNNs), particularly long short-term memory (LSTM) and gated recurrent units (GRUs), the study evaluates their effectiveness in predicting future cognitive states. These predictions have applications in real-time brain–computer interface (BCI) systems, enhancing responsiveness and adaptability in dynamic environments like robotic control. The proposed LSTM model demonstrated superior predictive accuracy for meditation states, achieving a Root Mean Squared Error (RMSE) of 10.90, while the GRU model excelled in predicting attention states, with an RMSE of 11.79. Both models outperformed the results provided by the proprietary eSense algorithm, reinforcing the potential of raw EEG data in cognitive-state analysis. Notably, inference times were optimized to under 50 milliseconds, making the models suitable for real-time applications. These findings underline the feasibility of using raw EEG signals from affordable devices for robust real-time prediction, offering a significant step forward in applied neuroscience. This research lays the groundwork for further exploration of RNN architectures in BCI applications, enabling safer, more intuitive, and personalized interactions in assistive technologies and beyond.
Background/Objectives: 3D printing has become an invaluable tool in medicine, enabling the creation of precise anatomical models for surgical planning and medical education. This study presents a comprehensive protocol for converting DICOM files into three-dimensional models and their subsequent transformation into GCODE files ready for 3D printing. Methods: We employed the open-source software “3D Slicer” for the initial conversion of the DICOM files, capitalising on its robust capabilities in segmentation and medical image processing. An optimised workflow was developed for the precise and efficient conversion of medical images into STL models, ensuring high fidelity in anatomical structures. The protocol was validated through three case studies, achieving elevated structural fidelity based on deviation analysis between the STL models and the original DICOM data. Furthermore, the segmentation process preserved morphological accuracy within a narrow deviation range, ensuring the reliable replication of anatomical features for medical applications. Our protocol provides an effective and accessible approach to generating 3D anatomical models with enhanced accuracy and reproducibility. In later stages, we utilised the “Ultimaker Cura” software to generate customised GCODE files tailored to the specifications of the 3D printer. Results: Our protocol offers an effective, accessible, and more accurate solution for creating 3D anatomical models from DICOM images. Furthermore, the versatility of this approach allows for its adaptation to various 3D printers and materials, expanding its utility in the medical and scientific community. Conclusions: This study presents a robust and reproducible approach for converting medical data into physical three-dimensional objects, paving the way for a wide range of applications in personalised medicine and advanced clinical practice. The selection of sample datasets from the 3D Slicer repository ensures standardisation and reproducibility, allowing for independent validation of the proposed workflow without ethical or logistical constraints related to patient data access. However, we acknowledge that future work could expand upon this by incorporating real patient datasets and benchmarking the protocol against alternative segmentation methods and software packages to further assess performance across different clinical scenarios. Essentially, this protocol can be particularly characterised by its commitment to open-source software and low-cost solutions, making advanced 3D modelling accessible to a wider audience. By leveraging open-access tools such as “3D Slicer” and “Ultimaker Cura”, we democratise the creation of anatomical models, ensuring that institutions with limited resources can also benefit from this technology, promoting innovation and inclusivity in medical sciences and education.
Neurodegenerative diseases present significant challenges in terms of mobility and autonomy for patients. In the current context of technological advances, brain–computer interfaces (BCIs) emerge as a promising tool to improve the quality of life of these patients. Therefore, in this study, we explore the feasibility of using low-cost commercial EEG headsets, such as Neurosky and Brainlink, for the control of robotic arms integrated into autonomous wheelchairs. These headbands, which offer attention and meditation values, have been adapted to provide intuitive control based on the eight EEG signal values read from Delta to Gamma (high and low/medium Gamma) collected from the users’ prefrontal area, using only two non-invasive electrodes. To ensure precise and adaptive control, we have incorporated a neural network that interprets these values in real time so that the response of the robotic arm matches the user’s intentions. The results suggest that this combination of BCIs, robotics, and machine learning techniques, such as neural networks, is not only technically feasible but also has the potential to radically transform the interaction of patients with neurodegenerative diseases with their environment.