Cultured epithelial autografts (CEA) are a life-saving option for massive burns when skin donor sites are insufficient for split thickness skin grafting (STSG). Long-term skin quality following grafting is usually judged with subjective scar scales or indirect biophysical testing with very limited standardized structural assessment potential. We report the use of optical coherence tomography (OCT) as a non-invasive imaging approach to objectively characterize long-term microstructure in a patient grafted with CEA. A 55-year-old man sustained 73% total body surface area (TBSA) burns with inhalation injury. Management included staged debridement and coverage with 64% TBSA STSG and 88% TBSA CEA, followed by prolonged intensive care and rehabilitation. Five years after injury, OCT imaging was performed using a commercial dynamic OCT device, with triplicate scans acquired at multiple anatomical sites representing full thickness skin grafting, meshed STSG at different expansion ratios, CEA alone, and combined meshed graft plus CEA. Control measurements were obtained from adjacent or contralateral healthy skin when available. Quantitative analysis included vascular plexus depth, vessel diameter, and vessel density, together with epidermal optical parameters, epidermal thickness, epidermal thickness variation, and surface roughness. OCT acquisition was feasible across all examined sites and was rapid and painless. Vascular measurements showed substantial intra-site variability and no consistent differences between grafted and healthy skin, suggesting an important influence of recipient site characteristics. In contrast, epidermal metrics portrayed variables across reconstruction strategies. Most grafted sites fell within the range of healthy skin for epidermal thickness, whereas areas treated with CEA alone displayed a markedly thinner epidermis 23.1 ± 2.5 µm and reduced epidermal thickness variation compared with healthy skin and meshed grafted sites; combined meshed graft plus CEA showed intermediate values. Surface roughness was generally comparable to healthy skin within standard deviation. These findings indicate that OCT can provide objective, rapid and non-invasive structural assessment of reconstructed skin years after burn. This case report also suggest that CEA might yield a persistent thin and uniform epidermal phenotype, warranting confirmation in larger prospective cohorts integrating OCT with functional and patient reported outcomes.
Human Epidermal Growth Factor Receptor 2 (HER2) status is a critical determinant of breast cancer prognosis and therapeutic strategy. Silver in situ Hybridization (SISH) enables HER2 gene assessment under bright-field microscopy while preserving tissue morphology, rendering it a clinically favoured alternative to fluorescence-based methods. Nevertheless, manual SISH slide interpretation is subjective, time-intensive, and susceptible to inter-observer variability, necessitating robust automated computational solutions. This work presents a systematic comparative evaluation of six deep learning architectures for multiclass HER2-SISH histopathology image classification across three diagnostic categories: Amplified, Non-Amplified, and ROI_Normal. The evaluated architectures span convolutional neural networks (MobileNetV3-Large, DenseNet121), transformer-based models (ViT-Small, DeiT-Small), and novel hybrid CNN-Transformer frameworks (DenseNet121-ViT, DenseNet121-DeiT). All models were trained under five-fold stratified cross-validation with a split-before-augmentation protocol and evaluated on an independent held-out test set over both the original and an augmented version of the publicly available HER2-SISH40x dataset. On the original dataset, DenseNet121 achieved the highest test accuracy of 82.10% with a macro F1-score of 77.33%. After offline rotational augmentation of the training partition, DenseNet121 achieved the best performance with test accuracy of 85.09% and macro F1-score of 80.61%, with ViT-Small (84.47%) and DenseNet121-ViT (83.23%) following closely. Per-class analysis revealed that ROI_Normal was consistently classified with the highest accuracy, whereas Amplified versus Non-Amplified discrimination remained the principal challenge across all evaluated models.
Accurate guidance for transbronchial pulmonary nodule biopsy requires visualisation of targets and tools beyond the airway wall, yet radial endobronchial ultrasound provides only a cross-sectional image orthogonal to the probe axis, and must be withdrawn before a tool is passed, leaving tissue acquisition unguided. Optical ultrasound (OpUS) imaging offers a miniaturised, fibre-based alternative that may enable image guidance from within the narrow distal bronchi. In this study, we evaluated the feasibility of OpUS imaging for endobronchial guidance by deploying a side-viewing, linearly actuated OpUS probe through a clinically available robotic bronchoscopy catheter, and assessed the corresponding imaging performance in separate bench-top experiments within an anatomically correct lung phantom containing airway structures and inclusions. Robotic integration was successful: the 1.8 mm outer-diameter OpUS probe could be delivered through the robotic catheter without visibly impairing catheter navigation, platform operation and probe delivery. The >2 m long probe maintained distal linear actuation through the catheter and tolerated catheter bending down to an estimated bend radius of approximately 5 mm before probe failure. The bench-top experiments confirmed how real-time OpUS imaging (4 Hz) could visualise the inclusion within the lung phantom from inside the airway lumen using a robotic bronchoscopy catheter. In addition, bench-top experiments on image guidance of needle insertions into phantoms confirmed that OpUS imaging is capable of clearly visualising the needle trajectory in settings matching the confined space of endobronchial interventions, supporting the great potential for OpUS image guidance of endobronchial lung biopsies.
Osmolyte-based Deep Eutectic Solvents (Osmo-DESs) inspired by biological systems are emerging as sustainable, biocompatible alternatives to conventional solvents for biomedical and pharmaceutical applications. In this study, we systematically investigate the thermophysical properties of multi-component Osmo-DESs based on naturally occurring metabolites, including amino-acids, sugars, sugar alcohols, and urea and trimethylamine N-oxide to mimic the complex intracellular environments found in nature. Using combinations of osmolytes, COSMO-RS computational screening and experimental validation employing polarized optical microscopy, Differential Scanning Calorimetry (DSC), and pH measurements we explore how increasing compositional complexity, from binary to multicomponent mixtures, including three-, four-, and six-component mixtures, influences thermophysical properties such as melting temperature (Tm), glass transition temperature (Tg), latent enthalpy of melting (ΔHm), and pH. Our results demonstrate that multi-component Osmo-DESs exhibit highly tunable thermal behaviour governed by intricate hydrogen-bond networks, with water fraction playing a pivotal role in modulating phase transitions and stability. Amino-acid based Osmo-DESs, such as Proline:Betaine and Proline:Sarcosine, show significant depression of melting temperatures of −30.09°C and −28.93°C respectively at low water mole fractions, are consistent with strong eutectic interactions. The addition of further components, including sugar alcohols and sugars, enables fine-tuning of both thermal and pH properties, with certain combinations maintaining alkaline pH even at high hydration levels. Among the binary TMAO-based Osmo-DESs, TMAO: Gua exhibited the lowest peak melting temperature of −46.13 °C at a water mole fraction of 0.71, whereas the multicomponent bio-inspired T-Osmo-DESs containing TMAO, betaine, taurine, and urea reached a minimum peak melting temperature of −43.78 °C at a water mole fraction of 0.61, closely mimicking the protective environments of living cells. By bridging computational predictions with experimental insights, this work advances the rational design of sustainable, green Osmo-DESs tailored for a variety of applications including biomolecular stabilization. Our findings highlight the versatility and promise of Osmo-DESs as next-generation solvents, offering new opportunities for environmentally friendly and application-specific solutions in biotechnology and medicine.
BackgroundTranscutaneous spinal neuromodulation (TSN) is the first therapy approved by US FDA and is delivered in conjunction with activity-based rehabilitation, but evidence describing routine clinical use across heterogeneous neurological populations remains limited.ObjectiveTo characterize the real-world implementation and documented safety of TSN delivered via xStep-assisted rehabilitation and to describe within-participant changes in routinely collected clinical outcomes.MethodsThis retrospective observational study reviewed de-identified records from 50 individuals with cerebral palsy (CP; n = 29), spinal cord injury (SCI; n = 11), or other neurological or structural conditions (n = 10). xStep was incorporated into individualized activity-based rehabilitation for approximately 1 h per session, 2–5 days per week. Outcomes available in routine records included hand-held dynamometry, Modified Ashworth Scale (MAS), International Standards for Neurological Classification of Spinal Cord Injury motor and sensory scores, 10-Meter Walk Test (10MWT), and Timed Up and Go (TUG) test. Analyses were descriptive; outcome-specific denominators are reported.ResultsDuring same-session dynamometry, mean force increased from 5.41 N without stimulation to 6.00 N with stimulation across 12 participants; 10 of 12 had a net increase in mean force. Spasticity decreased in at least one assessed muscle group in 13 of 24 participants. Among eight participants with SCI and paired assessments, mean motor score changed from 44.12 to 45.50 and mean sensory score from 124.62 to 129.12. Descriptive improvements were also observed in gait speed and TUG completion time among ambulatory participants, with variable assessment intervals and no untreated comparator. No device-related adverse event, serious adverse event, neurological deterioration, or device-related treatment discontinuation was documented in the reviewed records.ConclusionsxStep was integrated into individualized neurorehabilitation without documented device-related adverse events or treatment discontinuations in this retrospective cohort. The observed within-participant changes are clinical signals and do not establish efficacy or causality. Diagnosis-specific prospective studies with standardized exposure, predefined endpoints, and comparator conditions are needed.
