Designing a Digital Health Solution: A Platform for Automated Surveillance of Fungal Infection.

Surveillance of invasive fungal infection (IFI) requires laborious review of multiple sources of clinical information, while applying complex criteria to effectively identify relevant infections. These processes can be automated using artificial intelligence (AI) methodologies, including applying natural language processing (NLP) to clinical reports. However, developing a practically useful automated IFI surveillance tool requires consideration of the implementation context. We employed the Design Thinking Framework (DTF) to focus on the needs of end users of the tool to ensure sustained user engagement and enable its prospective validation. DTF allowed iterative generation of ideas and refinement of the final digital health solution. We believe this approach is key to increasing the likelihood that the solution will be implemented in clinical practice.

Detecting evidence of invasive fungal infections in cytology and histopathology reports enriched with concept-level annotations.

Invasive fungal infections (IFIs) are particularly dangerous to high-risk patients with haematological malignancies and are responsible for excessive mortality and delays in cancer therapy. Surveillance of IFI in clinical settings offers an opportunity to identify potential risk factors and evaluate new therapeutic strategies. However, manual surveillance is both time- and resource-intensive. As part of a broader project aimed to develop a system for automated IFI surveillance by leveraging electronic medical records, we present our approach to detecting evidence of IFI in the key diagnostic domain of histopathology. Using natural language processing (NLP), we analysed cytology and histopathology reports to identify IFI-positive reports. We compared a conventional bag-of-words classification model to a method that relies on concept-level annotations. Although the investment to prepare data supporting concept annotations is substantial, extracting targeted information specific to IFI as a pre-processing step increased the performance of the classifier from the PR AUC of 0.84 to 0.92 and enabled model interpretability. We have made publicly available the annotated dataset of 283 reports, the Cytology and Histopathology IFI Reports corpus (CHIFIR), to allow the clinical NLP research community to further build on our results.

Detection of self-harm and suicidal ideation in emergency department triage notes.

OBJECTIVE: Accurate identification of self-harm presentations to Emergency Departments (ED) can lead to more timely mental health support, aid in understanding the burden of suicidal intent in a population, and support impact evaluation of public health initiatives related to suicide prevention. Given lack of manual self-harm reporting in ED, we aim to develop an automated system for the detection of self-harm presentations directly from ED triage notes. MATERIALS AND METHODS: We frame this as supervised classification using natural language processing (NLP), utilizing a large data set of 477 627 free-text triage notes from ED presentations in 2012-2018 to The Royal Melbourne Hospital, Australia. The data were highly imbalanced, with only 1.4% of triage notes relating to self-harm. We explored various preprocessing techniques, including spelling correction, negation detection, bigram replacement, and clinical concept recognition, and several machine learning methods. RESULTS: Our results show that machine learning methods dramatically outperform keyword-based methods. We achieved the best results with a calibrated Gradient Boosting model, showing 90% Precision and 90% Recall (PR-AUC 0.87) on blind test data. Prospective validation of the model achieves similar results (88% Precision; 89% Recall). DISCUSSION: ED notes are noisy texts, and simple token-based models work best. Negation detection and concept recognition did not change the results while bigram replacement significantly impaired model performance. CONCLUSION: This first NLP-based classifier for self-harm in ED notes has practical value for identifying patients who would benefit from mental health follow-up in ED, and for supporting surveillance of self-harm and suicide prevention efforts in the population.

Chick Embryo Experimental Platform for Micrometastases Research in a 3D Tissue Engineering Model: Cancer Biology, Drug Development, and Nanotechnology Applications.

Colonization of distant organs by tumor cells is a critical step of cancer progression. The initial avascular stage of this process (micrometastasis) remains almost inaccessible to study due to the lack of relevant experimental approaches. Herein, we introduce an in vitro/in vivo model of organ-specific micrometastases of triple-negative breast cancer (TNBC) that is fully implemented in a cost-efficient chick embryo (CE) experimental platform. The model was built as three-dimensional (3D) tissue engineering constructs (TECs) combining human MDA-MB-231 cells and decellularized CE organ-specific scaffolds. TNBC cells colonized CE organ-specific scaffolds in 2-3 weeks, forming tissue-like structures. The feasibility of this methodology for basic cancer research, drug development, and nanomedicine was demonstrated on a model of hepatic micrometastasis of TNBC. We revealed that MDA-MB-231 differentially colonize parenchymal and stromal compartments of the liver-specific extracellular matrix (LS-ECM) and become more resistant to the treatment with molecular doxorubicin (Dox) and Dox-loaded mesoporous silica nanoparticles than in monolayer cultures. When grafted on CE chorioallantoic membrane, LS-ECM-based TECs induced angiogenic switch. These findings may have important implications for the diagnosis and treatment of TNBC. The methodology established here is scalable and adaptable for pharmacological testing and cancer biology research of various metastatic and primary tumors.

Machine learning reveals mesenchymal breast carcinoma cell adaptation in response to matrix stiffness.

Epithelial-mesenchymal transition (EMT) and its reverse process, mesenchymal-epithelial transition (MET), are believed to play key roles in facilitating the metastatic cascade. Metastatic lesions often exhibit a similar epithelial-like state to that of the primary tumour, in particular, by forming carcinoma cell clusters via E-cadherin-mediated junctional complexes. However, the factors enabling mesenchymal-like micrometastatic cells to resume growth and reacquire an epithelial phenotype in the target organ microenvironment remain elusive. In this study, we developed a workflow using image-based cell profiling and machine learning to examine morphological, contextual and molecular states of individual breast carcinoma cells (MDA-MB-231). MDA-MB-231 heterogeneous response to the host organ microenvironment was modelled by substrates with controllable stiffness varying from 0.2kPa (soft tissues) to 64kPa (bone tissues). We identified 3 distinct morphological cell types (morphs) varying from compact round-shaped to flattened irregular-shaped cells with lamellipodia, predominantly populating 2-kPa and >16kPa substrates, respectively. These observations were accompanied by significant changes in E-cadherin and vimentin expression. Furthermore, we demonstrate that the bone-mimicking substrate (64kPa) induced multicellular cluster formation accompanied by E-cadherin cell surface localisation. MDA-MB-231 cells responded to different substrate stiffness by morphological adaptation, changes in proliferation rate and cytoskeleton markers, and cluster formation on bone-mimicking substrate. Our results suggest that the stiffest microenvironment can induce MET.

Development of a Self-Harm Monitoring System for Victoria.

The prevention of suicide and suicide-related behaviour are key policy priorities in Australia and internationally. The World Health Organization has recommended that member states develop self-harm surveillance systems as part of their suicide prevention efforts. This is also a priority under Australia's Fifth National Mental Health and Suicide Prevention Plan. The aim of this paper is to describe the development of a state-based self-harm monitoring system in Victoria, Australia. In this system, data on all self-harm presentations are collected from eight hospital emergency departments in Victoria. A natural language processing classifier that uses machine learning to identify episodes of self-harm is currently being developed. This uses the free-text triage case notes, together with certain structured data fields, contained within the metadata of the incoming records. Post-processing is undertaken to identify primary mechanism of injury, substances consumed (including alcohol, illicit drugs and pharmaceutical preparations) and presence of psychiatric disorders. This system will ultimately leverage routinely collected data in combination with advanced artificial intelligence methods to support robust community-wide monitoring of self-harm. Once fully operational, this system will provide accurate and timely information on all presentations to participating emergency departments for self-harm, thereby providing a useful indicator for Australia's suicide prevention efforts.

Tracing upconversion nanoparticle penetration in human skin.

Due to their unique optical properties upconversion nanoparticles (UCNPs) provide exceptionally high contrast for imaging of true nanoparticle distribution in excised human skin. It makes possible to show penetration of solid nanoparticles in skin treated with chemical enhancers. We demonstrated tracing upconversion nanoparticles in excised human skin by means of optical microscopy at the discrete particle level sensitivity to obtain their penetration profiles, which was validated by laser-ablation inductively-coupled-plasma mass-spectrometry. To demonstrate utilities of our method, UCNPs were coated with polymers, formulated in water and chemical enhancers, and applied on excised human skin mounted on Franz cells, followed by imaging using a custom-built laser-scanning microscope. To evaluate the toxicity impact on skin by polymer-coated UCNPs, we introduced a tissue engineering model of viable epidermis made of decellularized chick embryo skin seeded with keratinocytes. UCNPs formulated in water stopped in stratum corneum, whereas UCNPs formulated in ethanol-water solution crossed stratum corneum and reached viable epidermis - hence, the enhancement effect for solid nanoparticles was detected by optical microscopy. All polymer-coated UCNPs were found nontoxic within the accepted safety levels. The keratinocyte resilience to polyethyleneimine-coated UCNPs was surprising considering cytotoxicity of polyethyleneimine to two-dimensional cell cultures.

Incoherent wavefront reconstruction by a retroemission device containing a thin fluorescent film: theory.

A retroemission device (REM) is an incoherent holographic device that represents a lenslet array situated on a substrate containing fluorescent material. Each lenslet focuses each wavelet of an optical wavefront incident on the REM device into a diffraction-limited volume (voxel) in the fluorescent material, so that the voxel coordinates encode the angle of incidence and curvature of the wavelet. The back-propagating fraction of the excited fluorescence is collected by the lenslet and quasi-collimated into a back-propagating wavelet. All wavelets are combined to reconstruct the incident wavefront propagating in the backward direction. We present a theoretical model of REM based on Fresnel-Kirchhoff approximation describing the reconstructed 3D image characteristics versus the thickness of the fluorescence film at the focal plane of the lenslets. Results of the computer simulations of the REM-based images of a point source, two axially separated point sources and an extended object (a circular rim) situated in the sagittal plane are presented. These results speak in favor of using a fluorescence film of minimum diffraction-limited thickness at the lenslet back focal plane. This REM structure minimizes the fluorescence background and improves the 3D imaging resolution in virtue of the exclusion of out-of-voxel fluorescence contributions to the reconstructed wavefront.