Precision Engineering of the Co-immobilization of Enzymes for Cascade Biocatalysis.

The design and orderly layered co-immobilization of multiple enzymes on resin particles remain challenging. In this study, the SpyTag/SpyCatcher binding pair was fused to the N-terminus of an alcohol dehydrogenase (ADH) and an aldo-keto reductase (AKR), respectively. A non-canonical amino acid (ncAA), p-azido-L-phenylalanine (p-AzF), as the anchor for covalent bonding enzymes, was genetically inserted into preselected sites in the AKR and ADH. Employing the two bioorthogonal counterparts of SpyTag/SpyCatcher and azide-alkyne cycloaddition for the immobilization of AKR and ADH enabled sequential dual-enzyme coating on porous microspheres. The ordered dual-enzyme reactor was subsequently used to synthesize (S)-1-(2-chlorophenyl)ethanol asymmetrically from the corresponding prochiral ketone, enabling the in situ regeneration of NADPH. The reactor exhibited a high catalytic conversion of 74 % and good reproducibility, retaining 80 % of its initial activity after six cycles. The product had 99.9 % ee, which that was maintained in each cycle. Additionally, the double-layer immobilization method significantly increased the enzyme loading capacity, which was approximately 1.7 times greater than that of traditional single-layer immobilization. More importantly, it simultaneously enabled both the purification and immobilization of multiple enzymes on carriers, thus providing a convenient approach to facilitate cascade biocatalysis.

Synthesis, characterization and irradiation enhances anticancer activity of liposome-loaded iridium(III) complexes.

Herein, we synthesized and characterized two novel iridium (III) complexes: [Ir(bzq)2(PPD)](PF6) (4a, with bzq = deprotonated benzo[h]quinoline and PPD = pteridino[6,7-f][1,10]phenanthroline-11,13-diamine) and [Ir(piq)2(PPD)](PF6) (4b, with piq = deprotonated 1-phenylisoquinoline). The anticancer efficacy of these complexes, 4a and 4b, was investigated using 3-(4,5-dimethylthiazole)-2,5-diphenltetraazolium bromide (MTT). Complex 4a exhibited no cytotoxic activity, while 4b demonstrated moderate efficacy against SGC-7901, A549, and HepG2 cancer cells. To enhance their anticancer potential, we explored two strategies: (I) light irradiation and (II) encapsulation of the complexes in liposomes, resulting in the formation of 4alip and 4blip. Both strategies significantly increased the ability of 4a, 4b to kill cancer cells. The cellular studies indicated that both the free complexes 4a, 4b and their liposomal forms 4alip and 4blip effectively inhibited cell proliferation. The cell cycle arrest analysis uncovered 4alip and 4blip arresting cell growth in the S period. Additionally, we investigated apoptosis and ferroptosis pathways, observing an increase in malondialdehyde (MDA) levels, a reduction of glutathione (GSH), a down-regulation of GPX4 (glutathione peroxidase) expression, and lipid peroxidation. The effects on mitochondrial membrane potential and intracellular Ca2+ concentrations were also examined, revealing that both light-activated and liposomal forms of 4alip and 4blip caused a decline in mitochondrial membrane potential and an enhancement in intracellular Ca2+ levels. In conclusion, these complexes and them encapsulated liposomes induce cell death through apoptosis and ferroptosis.

Near-infrared light-driven asymmetric photolytic reduction of ketone using inorganic-enzyme hybrid biocatalyst.

Effective photolytic regeneration of the NAD(P)H cofactor in enzymatic reductions is an important and elusive goal in biocatalysis. It can, in principle, be achieved using a near-infrared light (NIR) driven artificial photosynthesis system employing H2O as the sacrificial reductant. To this end we utilized TiO2/reduced graphene quantum dots (r-GQDs), combined with a novel rhodium electron mediator, to continuously supply NADPH in situ for aldo-keto reductase (AKR) mediated asymmetric reductions under NIR irradiation. This upconversion system, in which the Ti-O-C bonds formed between r-GQDs and TiO2 enabled efficient interfacial charge transfer, was able to regenerate NADPH efficiently in 64 % yield in 105 min. Based on this, the pharmaceutical intermediate (R)-1-(3,5-bis(trifluoromethyl)phenyl)ethan-1-ol was obtained, in 84 % yield and 99.98 % ee, by reduction of the corresponding ketone. The photo-enzymatic system is recyclable with a polymeric electron mediator, which maintained 66 % of its original catalytic efficiency and excellent enantioselectivity (99.9 % ee) after 6 cycles.

An ASER AI/ML expert panel formative user research study for an interpretable interactive splenic AAST grading graphical user interface prototype.

PURPOSE: The AAST Organ Injury Scale is widely adopted for splenic injury severity but suffers from only moderate inter-rater agreement. This work assesses SpleenPro, a prototype interactive explainable artificial intelligence/machine learning (AI/ML) diagnostic aid to support AAST grading, for effects on radiologist dwell time, agreement, clinical utility, and user acceptance. METHODS: Two trauma radiology ad hoc expert panelists independently performed timed AAST grading on 76 admission CT studies with blunt splenic injury, first without AI/ML assistance, and after a 2-month washout period and randomization, with AI/ML assistance. To evaluate user acceptance, three versions of the SpleenPro user interface with increasing explainability were presented to four independent expert panelists with four example cases each. A structured interview consisting of Likert scales and free responses was conducted, with specific questions regarding dimensions of diagnostic utility (DU); mental support (MS); effort, workload, and frustration (EWF); trust and reliability (TR); and likelihood of future use (LFU). RESULTS: SpleenPro significantly decreased interpretation times for both raters. Weighted Cohen's kappa increased from 0.53 to 0.70 with AI/ML assistance. During user acceptance interviews, increasing explainability was associated with improvement in Likert scores for MS, EWF, TR, and LFU. Expert panelists indicated the need for a combined early notification and grading functionality, PACS integration, and report autopopulation to improve DU. CONCLUSIONS: SpleenPro was useful for improving objectivity of AAST grading and increasing mental support. Formative user research identified generalizable concepts including the need for a combined detection and grading pipeline and integration with the clinical workflow.

Greater target or lure variability? An exploration on the effects of stimulus types and memory paradigms.

In recognition memory, the variance of the target distribution is almost universally found to be greater than that of the lure distribution. However, these estimates commonly come from long-term memory paradigms where words are used as stimuli. Two exceptions to this rule have found evidence for greater lure variability: a short-term memory task (Yotsumoto et al., Memory & Cognition, 36, 282-294 2008) and in an eyewitness memory paradigm (Wixted et al., Cognitive Psychology, 105, 81-114 2018). In the present work, we conducted a series of recognition memory experiments using different stimulus (faces vs. words) along with different paradigms (long-term vs. short-term paradigms) to evaluate whether either of these conditions would result in greater variability in lure items. Greater target variability was observed across stimulus types and memory paradigms. This suggests that factors other than stimuli and retention interval might be responsible for cases where variability is less for targets than lures.

A vendor-agnostic, PACS integrated, and DICOM-compatible software-server pipeline for testing segmentation algorithms within the clinical radiology workflow.

Background: Reproducible approaches are needed to bring AI/ML for medical image analysis closer to the bedside. Investigators wishing to shadow test cross-sectional medical imaging segmentation algorithms on new studies in real-time will benefit from simple tools that integrate PACS with on-premises image processing, allowing visualization of DICOM-compatible segmentation results and volumetric data at the radiology workstation. Purpose: In this work, we develop and release a simple containerized and easily deployable pipeline for shadow testing of segmentation algorithms within the clinical workflow. Methods: Our end-to-end automated pipeline has two major components- 1. A router/listener and anonymizer and an OHIF web viewer backstopped by a DCM4CHEE DICOM query/retrieve archive deployed in the virtual infrastructure of our secure hospital intranet, and 2. An on-premises single GPU workstation host for DICOM/NIfTI conversion steps, and image processing. DICOM images are visualized in OHIF along with their segmentation masks and associated volumetry measurements (in mL) using DICOM SEG and structured report (SR) elements. Since nnU-net has emerged as a widely-used out-of-the-box method for training segmentation models with state-of-the-art performance, feasibility of our pipleine is demonstrated by recording clock times for a traumatic pelvic hematoma nnU-net model. Results: Mean total clock time from PACS send by user to completion of transfer to the DCM4CHEE query/retrieve archive was 5 min 32 s (± SD of 1 min 26 s). This compares favorably to the report turnaround times for whole-body CT exams, which often exceed 30 min, and illustrates feasibility in the clinical setting where quantitative results would be expected prior to report sign-off. Inference times accounted for most of the total clock time, ranging from 2 min 41 s to 8 min 27 s. All other virtual and on-premises host steps combined ranged from a minimum of 34 s to a maximum of 48 s. Conclusion: The software worked seamlessly with an existing PACS and could be used for deployment of DL models within the radiology workflow for prospective testing on newly scanned patients. Once configured, the pipeline is executed through one command using a single shell script. The code is made publicly available through an open-source license at "https://github.com/vastc/," and includes a readme file providing pipeline config instructions for host names, series filter, other parameters, and citation instructions for this work.

Accelerating voxelwise annotation of cross-sectional imaging through AI collaborative labeling with quality assurance and bias mitigation.

Background: precision-medicine quantitative tools for cross-sectional imaging require painstaking labeling of targets that vary considerably in volume, prohibiting scaling of data annotation efforts and supervised training to large datasets for robust and generalizable clinical performance. A straight-forward time-saving strategy involves manual editing of AI-generated labels, which we call AI-collaborative labeling (AICL). Factors affecting the efficacy and utility of such an approach are unknown. Reduction in time effort is not well documented. Further, edited AI labels may be prone to automation bias. Purpose: In this pilot, using a cohort of CTs with intracavitary hemorrhage, we evaluate both time savings and AICL label quality and propose criteria that must be met for using AICL annotations as a high-throughput, high-quality ground truth. Methods: 57 CT scans of patients with traumatic intracavitary hemorrhage were included. No participant recruited for this study had previously interpreted the scans. nnU-net models trained on small existing datasets for each feature (hemothorax/hemoperitoneum/pelvic hematoma; n = 77-253) were used in inference. Two common scenarios served as baseline comparison- de novo expert manual labeling, and expert edits of trained staff labels. Parameters included time effort and image quality graded by a blinded independent expert using a 9-point scale. The observer also attempted to discriminate AICL and expert labels in a random subset (n = 18). Data were compared with ANOVA and post-hoc paired signed rank tests with Bonferroni correction. Results: AICL reduced time effort 2.8-fold compared to staff label editing, and 8.7-fold compared to expert labeling (corrected p < 0.0006). Mean Likert grades for AICL (8.4, SD:0.6) were significantly higher than for expert labels (7.8, SD:0.9) and edited staff labels (7.7, SD:0.8) (corrected p < 0.0006). The independent observer failed to correctly discriminate AI and human labels. Conclusion: For our use case and annotators, AICL facilitates rapid large-scale curation of high-quality ground truth. The proposed quality control regime can be employed by other investigators prior to embarking on AICL for segmentation tasks in large datasets.

A vendor-agnostic, PACS integrated, and DICOMcompatible software-server pipeline for testing segmentation algorithms within the clinical radiology workflow.

Background: Reproducible approaches are needed to bring AI/ML for medical image analysis closer to the bedside. Investigators wishing to shadow test cross-sectional medical imaging segmentation algorithms on new studies in real-time will benefit from simple tools that integrate PACS with on-premises image processing, allowing visualization of DICOM-compatible segmentation results and volumetric data at the radiology workstation. Purpose: In this work, we develop and release a simple containerized and easily deployable pipeline for shadow testing of segmentation algorithms within the clinical workflow. Methods: Our end-to-end automated pipeline has two major components-1. a router/listener and anonymizer and an OHIF web viewer backstopped by a DCM4CHEE DICOM query/retrieve archive deployed in the virtual infrastructure of our secure hospital intranet, and 2. An on-premises single GPU workstation host for DICOM/NIfTI conversion steps, and image processing. DICOM images are visualized in OHIF along with their segmentation masks and associated volumetry measurements (in mL) using DICOM SEG and structured report (SR) elements. Feasibility is demonstrated by recording clock times for a traumatic pelvic hematoma cascaded nnU-net model. Results: Mean total clock time from PACS send by user to completion of transfer to the DCM4CHEE query/retrieve archive was 5 minutes 32 seconds (+/- SD of 1 min 26 sec). This compares favorably to the report turnaround times for whole-body CT exams, which often exceed 30 minutes. Inference times accounted for most of the total clock time, ranging from 2 minutes 41 seconds to 8 minutes 27 seconds. All other virtual and on-premises host steps combined ranged from a minimum of 34 seconds to a maximum of 48 seconds. Conclusion: The software worked seamlessly with an existing PACS and could be used for deployment of DL models within the radiology workflow for prospective testing on newly scanned patients. Once configured, the pipeline is executed through one command using a single shell script. The code is made publicly available through an open-source license at "https://github.com/vastc/", and includes a readme file providing pipeline config instructions for host names, series filter, other parameters, and citation instructions for this work.

Engineering ketoreductases for the enantioselective synthesis of chiral alcohols.

The use of engineered ketoreductases (KREDS), both as whole microbial cells and isolated enzymes, in the highly enantiospecific reduction of prochiral ketones is reviewed. The homochiral alcohol products are key intermediates in, for example, pharmaceuticals synthesis. The application of sophisticated protein engineering and enzyme immobilisation techniques to increase industrial viability are discussed.

Mechanochromic Optical/Electrical Skin for Ultrasensitive Dual-Signal Sensing.

Following earlier research efforts dedicated to the realization of multifunctional sensing, recent developments of artificial skins endeavor to go beyond human sensory functions by integrating interactive visualization of strain and pressure stimuli. Inspired by the microcracked structure of spider slit organs and the mechanochromic mechanism of chameleons, this work aims to design a flexible optical/electrical skin (OE-skin) capable of responding to complex stimuli with interactive feedback of human-readable structural colors. The OE-skin consists of an ionic electrode combined with an elastomer dielectric layer, a chromotropic layer containing photonic crystals and a conductive carbon nanotube/MXene layer. The electrode/dielectric layers function as a capacitive pressure sensor. The mechanochromic photonic crystals of ferroferric oxide-carbon magnetic arrays embedded in the gelatin/polyacrylamide stretchable hydrogel film perceive strain and pressure stimuli with bright color switching outputs in the full visible spectrum. The underlying microcracked conductive layer is devoted to ultrasensitive strain sensing with a gauge factor of 191.8. The multilayered OE-skin delivers an ultrafast, accurate response for capacitive pressure sensing with a detection limit of 75 Pa and long-term stability of 5000 cycles, while visualizing complex deformations in the form of high-resolution spatial colors. These findings offer deep insights into the rational design of OE-skins as multifunctional sensing devices.

Direct Gene Expression Profile Prediction for Uveal Melanoma from Digital Cytopathology Images via Deep Learning and Salient Image Region Identification.

Objective: To demonstrate that deep learning (DL) methods can produce robust prediction of gene expression profile (GEP) in uveal melanoma (UM) based on digital cytopathology images. Design: Evaluation of a diagnostic test or technology. Subjects Participants and Controls: Deidentified smeared cytology slides stained with hematoxylin and eosin obtained from a fine needle aspirated from UM. Methods: Digital whole-slide images were generated by fine-needle aspiration biopsies of UM tumors that underwent GEP testing. A multistage DL system was developed with automatic region-of-interest (ROI) extraction from digital cytopathology images, an attention-based neural network, ROI feature aggregation, and slide-level data augmentation. Main Outcome Measures: The ability of our DL system in predicting GEP on a slide (patient) level. Data were partitioned at the patient level (73% training; 27% testing). Results: In total, our study included 89 whole-slide images from 82 patients and 121 388 unique ROIs. The testing set included 24 slides from 24 patients (12 class 1 tumors; 12 class 2 tumors; 1 slide per patient). Our DL system for GEP prediction achieved an area under the receiver operating characteristic curve of 0.944, an accuracy of 91.7%, a sensitivity of 91.7%, and a specificity of 91.7% on a slide-level analysis. The incorporation of slide-level feature aggregation and data augmentation produced a more predictive DL model (P = 0.0031). Conclusions: Our current work established a complete pipeline for GEP prediction in UM tumors: from automatic ROI extraction from digital cytopathology whole-slide images to slide-level predictions. Our DL system demonstrated robust performance and, if validated prospectively, could serve as an image-based alternative to GEP testing.

Toward automated interpretable AAST grading for blunt splenic injury.

BACKGROUND: The American Association for the Surgery of Trauma (AAST) splenic organ injury scale (OIS) is the most frequently used CT-based grading system for blunt splenic trauma. However, reported inter-rater agreement is modest, and an algorithm that objectively automates grading based on transparent and verifiable criteria could serve as a high-trust diagnostic aid. PURPOSE: To pilot the development of an automated interpretable multi-stage deep learning-based system to predict AAST grade from admission trauma CT. METHODS: Our pipeline includes 4 parts: (1) automated splenic localization, (2) Faster R-CNN-based detection of pseudoaneurysms (PSA) and active bleeds (AB), (3) nnU-Net segmentation and quantification of splenic parenchymal disruption (SPD), and (4) a directed graph that infers AAST grades from detection and segmentation results. Training and validation is performed on a dataset of adult patients (age ≥ 18) with voxelwise labeling, consensus AAST grading, and hemorrhage-related outcome data (n = 174). RESULTS: AAST classification agreement (weighted κ) between automated and consensus AAST grades was substantial (0.79). High-grade (IV and V) injuries were predicted with accuracy, positive predictive value, and negative predictive value of 92%, 95%, and 89%. The area under the curve for predicting hemorrhage control intervention was comparable between expert consensus and automated AAST grading (0.83 vs 0.88). The mean combined inference time for the pipeline was 96.9 s. CONCLUSIONS: The results of our method were rapid and verifiable, with high agreement between automated and expert consensus grades. Diagnosis of high-grade lesions and prediction of hemorrhage control intervention produced accurate results in adult patients.

Explainable medical imaging AI needs human-centered design: guidelines and evidence from a systematic review.

Transparency in Machine Learning (ML), often also referred to as interpretability or explainability, attempts to reveal the working mechanisms of complex models. From a human-centered design perspective, transparency is not a property of the ML model but an affordance, i.e., a relationship between algorithm and users. Thus, prototyping and user evaluations are critical to attaining solutions that afford transparency. Following human-centered design principles in highly specialized and high stakes domains, such as medical image analysis, is challenging due to the limited access to end users and the knowledge imbalance between those users and ML designers. To investigate the state of transparent ML in medical image analysis, we conducted a systematic review of the literature from 2012 to 2021 in PubMed, EMBASE, and Compendex databases. We identified 2508 records and 68 articles met the inclusion criteria. Current techniques in transparent ML are dominated by computational feasibility and barely consider end users, e.g. clinical stakeholders. Despite the different roles and knowledge of ML developers and end users, no study reported formative user research to inform the design and development of transparent ML models. Only a few studies validated transparency claims through empirical user evaluations. These shortcomings put contemporary research on transparent ML at risk of being incomprehensible to users, and thus, clinically irrelevant. To alleviate these shortcomings in forthcoming research, we introduce the INTRPRT guideline, a design directive for transparent ML systems in medical image analysis. The INTRPRT guideline suggests human-centered design principles, recommending formative user research as the first step to understand user needs and domain requirements. Following these guidelines increases the likelihood that the algorithms afford transparency and enable stakeholders to capitalize on the benefits of transparent ML.

A pilot study of deep learning-based CT volumetry for traumatic hemothorax.

PURPOSE: We employ nnU-Net, a state-of-the-art self-configuring deep learning-based semantic segmentation method for quantitative visualization of hemothorax (HTX) in trauma patients, and assess performance using a combination of overlap and volume-based metrics. The accuracy of hemothorax volumes for predicting a composite of hemorrhage-related outcomes - massive transfusion (MT) and in-hospital mortality (IHM) not related to traumatic brain injury - is assessed and compared to subjective expert consensus grading by an experienced chest and emergency radiologist. MATERIALS AND METHODS: The study included manually labeled admission chest CTs from 77 consecutive adult patients with non-negligible (≥ 50 mL) traumatic HTX between 2016 and 2018 from one trauma center. DL results of ensembled nnU-Net were determined from fivefold cross-validation and compared to individual 2D, 3D, and cascaded 3D nnU-Net results using the Dice similarity coefficient (DSC) and volume similarity index. Pearson's r, intraclass correlation coefficient (ICC), and mean bias were also determined for the best performing model. Manual and automated hemothorax volumes and subjective hemothorax volume grades were analyzed as predictors of MT and IHM using AUC comparison. Volume cut-offs yielding sensitivity or specificity ≥ 90% were determined from ROC analysis. RESULTS: Ensembled nnU-Net achieved a mean DSC of 0.75 (SD: ± 0.12), and mean volume similarity of 0.91 (SD: ± 0.10), Pearson r of 0.93, and ICC of 0.92. Mean overmeasurement bias was only 1.7 mL despite a range of manual HTX volumes from 35 to 1503 mL (median: 178 mL). AUC of automated volumes for the composite outcome was 0.74 (95%CI: 0.58-0.91), compared to 0.76 (95%CI: 0.58-0.93) for manual volumes, and 0.76 (95%CI: 0.62-0.90) for consensus expert grading (p = 0.93). Automated volume cut-offs of 77 mL and 334 mL predicted the outcome with 93% sensitivity and 90% specificity respectively. CONCLUSION: Automated HTX volumetry had high method validity, yielded interpretable visual results, and had similar performance for the hemorrhage-related outcomes assessed compared to manual volumes and expert consensus grading. The results suggest promising avenues for automated HTX volumetry in research and clinical care.

Programmable Polyproteams of Tyrosine Ammonia Lyases as Cross-Linked Enzymes for Synthesizing p-Coumaric Acid.

Ideal immobilization with enhanced biocatalyst activity and thermostability enables natural enzymes to serve as a powerful tool to yield synthetically useful chemicals in industry. Such an enzymatic method strategy becomes easier and more convenient with the use of genetic and protein engineering. Here, we developed a covalent programmable polyproteam of tyrosine ammonia lyases (TAL-CLEs) by fusing SpyTag and SpyCatcher peptides into the N-terminal and C-terminal of the TAL, respectively. The resulting circular enzymes were clear after the spontaneous isopeptide bonds formed between the SpyTag and SpyCatcher. Furthermore, the catalytic performance of the TAL-CLEs was measured via a synthesis sample of p-Coumaric acid. Our TAL-CLEs showed excellent catalytic efficiency, with 98.31 ± 1.14% yield of the target product-which is 4.15 ± 0.08 times higher than that of traditional glutaraldehyde-mediated enzyme aggregates. They also showed over four times as much enzyme-activity as wild-type TAL does and demonstrated good reusability, and so may become a good candidate for industrial enzymes.

Conductive Hydrogel Conduits with Growth Factor Gradients for Peripheral Nerve Repair in Diabetics with Non-Suture Tape.

Diabetic patients suffer from peripheral nerve injury with slow and incomplete regeneration owing to hyperglycemia and microvascular complications. This study develops a graphene-based nerve guidance conduit by incorporating natural double network hydrogel and a neurotrophic concentration gradient with non-invasive treatment for diabetics. GelMA/silk fibroin double network hydrogel plays quadruple roles for rapid setting/curing, suitable mechanical supporting, good biocompatibility, and sustainable growth factor delivery. Meanwhile, graphene mesh can improve the toughness of conduit and enhance conductivity of conduit for regeneration. Here, novel silk tapes show quick and tough adhesion of wet tissue by dual mechanism to replace suture step. The in vivo results demonstrate that gradient concentration of netrin-1 in conduit have better performance than uniform concentration caused by chemotaxis phenomenon for axon extension, remyelination, and angiogenesis. Altogether, GelMA/silk graphene conduit with gradient netrin-1 and dry double-sided adhesive tape can significantly promote repairing of peripheral nerve injury and inhibit the atrophy of muscles for diabetics.

Interdigitated Three-Dimensional Heterogeneous Nanocomposites for High-Performance Mechanochromic Smart Membranes.

Mechanochromic smart membranes capable of optical modulation have great potential in smart windows, artificial skins, and camouflage. However, the realization of high-contrast optical modulation based on light scattering activated at a low strain remains challenging. Here, we present a strategy for designing mechanochromic scattering membranes by introducing a Young's modulus mismatch between the two interdigitated polydimethylsiloxane phases with weak interfaces in a periodic three-dimensional (3D) structure. The refractive index-matched interfaces of the nanocomposite provide a high optical transparency of 93%. Experimental and computational studies reveal that the 3D heterogeneity facilitates the generation of numerous nanoscale debonds or "nanogaps" at the modulus-mismatching interfaces, enabling incident light scattering under tension. The heterogeneous scatterer delivers both a high transmittance contrast of >50% achieved at 15% strain and a maximum contrast of 82%. When used as a smart window, the membrane demonstrates effective diffusion of transmitting sunlight, leading to moderate indoor illumination by eliminating extremely bright or dark spots. At the other extreme, such a 3D heterogeneous design with strongly bonded interfaces can enhance the coloration sensitivity of mechanophore-dyed nanocomposites. This work presents insights into the design principles of advanced mechanochromic smart membranes.

Flexible temperature sensors made of aligned electrospun carbon nanofiber films with outstanding sensitivity and selectivity towards temperature.

Continuous real-time measurement of body temperature using a wearable sensor is an essential part of human health monitoring. Electrospun aligned carbon nanofiber (ACNF) films are employed to assemble flexible temperature sensors. The temperature sensor prepared at a low carbonization temperature of 650 °C yields an outstanding sensitivity of 1.52% °C-1, high accuracy, good linearity, fast response time and excellent long-term durability. Moreover, it exhibits high discriminability towards temperature amidst other unwanted stimuli and maintains its original performance even after repeated stretch/release cycles because of highly-aligned structures. The correlation between the atomic structure and the temperature sensing performance of ACNF sensors is established. Contrary to conventional highly conductive temperature sensors, the ACNF sensor with a low electrical conductivity prepared at a low carbonization temperature ameliorates the temperature sensing performance. This anomaly is explained by (i) the smaller and more disordered sp2 carbon crystallites yielding a high negative temperature coefficient, (ii) a larger number of defects, and (iii) a higher pyridinic-N content generating abundant entrapped and localized electrons which are activated once sufficient thermal energy is available. Flexible ACNF sensor's overall performance is among the best-known carbon material-based flexible temperature sensors, demonstrating potential applications in emerging healthcare and flexible electronics technologies.

Anisotropic, Wrinkled, and Crack-Bridging Structure for Ultrasensitive, Highly Selective Multidirectional Strain Sensors.

Flexible multidirectional strain sensors are crucial to accurately determining the complex strain states involved in emerging sensing applications. Although considerable efforts have been made to construct anisotropic structures for improved selective sensing capabilities, existing anisotropic sensors suffer from a trade-off between high sensitivity and high stretchability with acceptable linearity. Here, an ultrasensitive, highly selective multidirectional sensor is developed by rational design of functionally different anisotropic layers. The bilayer sensor consists of an aligned carbon nanotube (CNT) array assembled on top of a periodically wrinkled and cracked CNT-graphene oxide film. The transversely aligned CNT layer bridge the underlying longitudinal microcracks to effectively discourage their propagation even when highly stretched, leading to superior sensitivity with a gauge factor of 287.6 across a broad linear working range of up to 100% strain. The wrinkles generated through a pre-straining/releasing routine in the direction transverse to CNT alignment is responsible for exceptional selectivity of 6.3, to the benefit of accurate detection of loading directions by the multidirectional sensor. This work proposes a unique approach to leveraging the inherent merits of two cross-influential anisotropic structures to resolve the trade-off among sensitivity, selectivity, and stretchability, demonstrating promising applications in full-range, multi-axis human motion detection for wearable electronics and smart robotics.

Deep hiearchical multi-label classification applied to chest X-ray abnormality taxonomies.

Chest X-rays (CXRs) are a crucial and extraordinarily common diagnostic tool, leading to heavy research for computer-aided diagnosis (CAD) solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep hierarchical multi-label classification (HMLC) approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. Finally, we demonstrate that HMLC can be an effective means to manage missing or incomplete labels. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting abnormality labels from the CXR arm of the Prostate, Lung, Colorectal and Ovarian (PLCO) dataset, which comprises over 198,000 manually annotated CXRs. When using complete labels, we report a mean area under the curve (AUC) of 0.887, the highest yet reported for this dataset. These results are supported by ancillary experiments on the PadChest dataset, where we also report significant improvements, 1.2% and 4.1% in AUC and average precision, respectively over strong "flat" classifiers. Finally, we demonstrate that our HMLC approach can much better handle incompletely labelled data. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.