Automated reporting of cervical biopsies using artificial intelligence.

When detected at an early stage, the 5-year survival rate for people with invasive cervical cancer is 92%. Being aware of signs and symptoms of cervical cancer and early detection greatly improve the chances of successful treatment. We have developed an Artificial Intelligence (AI) algorithm, trained and evaluated on cervical biopsies for automated reporting of digital diagnostics. The aim is to increase overall efficiency of pathological diagnosis and to have the performance tuned to high sensitivity for malignant cases. Having a tool for triage/identifying cancer and high grade lesions may potentially reduce reporting time by identifying areas of interest in a slide for the pathologist and therefore improving efficiency. We trained and validated our algorithm on 1738 cervical WSIs with one WSI per patient. On the independent test set of 811 WSIs, we achieved 93.4% malignant sensitivity for classifying slides. Recognising a WSI, with our algorithm, takes approximately 1.5 minutes on the NVIDIA Tesla V100 GPU. Whole slide images of different formats (TIFF, iSyntax, and CZI) can be processed using this code, and it is easily extendable to other formats.

Classification of hyper-scale multimodal imaging datasets.

Algorithms that classify hyper-scale multi-modal datasets, comprising of millions of images, into constituent modality types can help researchers quickly retrieve and classify diagnostic imaging data, accelerating clinical outcomes. This research aims to demonstrate that a deep neural network that is trained on a hyper-scale dataset (4.5 million images) composed of heterogeneous multi-modal data can be used to obtain significant modality classification accuracy (96%). By combining 102 medical imaging datasets, a dataset of 4.5 million images was created. A ResNet-50, ResNet-18, and VGG16 were trained to classify these images by the imaging modality used to capture them (Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and X-ray) across many body locations. The classification accuracy of the models was then tested on unseen data. The best performing model achieved classification accuracy of 96% on unseen data, which is on-par, or exceeds the accuracy of more complex implementations using EfficientNets or Vision Transformers (ViTs). The model achieved a balanced accuracy of 86%. This research shows it is possible to train Deep Learning (DL) Convolutional Neural Networks (CNNs) with hyper-scale multimodal datasets, composed of millions of images. Such models can find use in real-world applications with volumes of image data in the hyper-scale range, such as medical imaging repositories, or national healthcare institutions. Further research can expand this classification capability to include 3D-scans.

A divide and conquer approach to maximise deep learning mammography classification accuracies.

Breast cancer claims 11,400 lives on average every year in the UK, making it one of the deadliest diseases. Mammography is the gold standard for detecting early signs of breast cancer, which can help cure the disease during its early stages. However, incorrect mammography diagnoses are common and may harm patients through unnecessary treatments and operations (or a lack of treatment). Therefore, systems that can learn to detect breast cancer on their own could help reduce the number of incorrect interpretations and missed cases. Various deep learning techniques, which can be used to implement a system that learns how to detect instances of breast cancer in mammograms, are explored throughout this paper. Convolution Neural Networks (CNNs) are used as part of a pipeline based on deep learning techniques. A divide and conquer approach is followed to analyse the effects on performance and efficiency when utilising diverse deep learning techniques such as varying network architectures (VGG19, ResNet50, InceptionV3, DenseNet121, MobileNetV2), class weights, input sizes, image ratios, pre-processing techniques, transfer learning, dropout rates, and types of mammogram projections. This approach serves as a starting point for model development of mammography classification tasks. Practitioners can benefit from this work by using the divide and conquer results to select the most suitable deep learning techniques for their case out-of-the-box, thus reducing the need for extensive exploratory experimentation. Multiple techniques are found to provide accuracy gains relative to a general baseline (VGG19 model using uncropped 512 × 512 pixels input images with a dropout rate of 0.2 and a learning rate of 1 × 10-3) on the Curated Breast Imaging Subset of DDSM (CBIS-DDSM) dataset. These techniques involve transfer learning pre-trained ImagetNet weights to a MobileNetV2 architecture, with pre-trained weights from a binarised version of the mini Mammography Image Analysis Society (mini-MIAS) dataset applied to the fully connected layers of the model, coupled with using weights to alleviate class imbalance, and splitting CBIS-DDSM samples between images of masses and calcifications. Using these techniques, a 5.6% gain in accuracy over the baseline model was accomplished. Other deep learning techniques from the divide and conquer approach, such as larger image sizes, do not yield increased accuracies without the use of image pre-processing techniques such as Gaussian filtering, histogram equalisation and input cropping.

Detection of malignancy in whole slide images of endometrial cancer biopsies using artificial intelligence.

In this study we use artificial intelligence (AI) to categorise endometrial biopsy whole slide images (WSI) from digital pathology as either "malignant", "other or benign" or "insufficient". An endometrial biopsy is a key step in diagnosis of endometrial cancer, biopsies are viewed and diagnosed by pathologists. Pathology is increasingly digitised, with slides viewed as images on screens rather than through the lens of a microscope. The availability of these images is driving automation via the application of AI. A model that classifies slides in the manner proposed would allow prioritisation of these slides for pathologist review and hence reduce time to diagnosis for patients with cancer. Previous studies using AI on endometrial biopsies have examined slightly different tasks, for example using images alongside genomic data to differentiate between cancer subtypes. We took 2909 slides with "malignant" and "other or benign" areas annotated by pathologists. A fully supervised convolutional neural network (CNN) model was trained to calculate the probability of a patch from the slide being "malignant" or "other or benign". Heatmaps of all the patches on each slide were then produced to show malignant areas. These heatmaps were used to train a slide classification model to give the final slide categorisation as either "malignant", "other or benign" or "insufficient". The final model was able to accurately classify 90% of all slides correctly and 97% of slides in the malignant class; this accuracy is good enough to allow prioritisation of pathologists' workload.

Weakly supervised learning and interpretability for endometrial whole slide image diagnosis.

Fully supervised learning for whole slide image-based diagnostic tasks in histopathology is problematic due to the requirement for costly and time-consuming manual annotation by experts. Weakly supervised learning that utilizes only slide-level labels during training is becoming more widespread as it relieves this burden, but has not yet been applied to endometrial whole slide images, in iSyntax format. In this work, we apply a weakly supervised learning algorithm to a real-world dataset of this type for the first time, with over 85% validation accuracy and over 87% test accuracy. We then employ interpretability methods including attention heatmapping, feature visualization, and a novel end-to-end saliency-mapping approach to identify distinct morphologies learned by the model and build an understanding of its behavior. These interpretability methods, alongside consultation with expert pathologists, allow us to make comparisons between machine-learned knowledge and consensus in the field. This work contributes to the state of the art by demonstrating a robust practical application of weakly supervised learning on a real-world digital pathology dataset and shows the importance of fine-grained interpretability to support understanding and evaluation of model performance in this high-stakes use case.

Automated Remote Pulse Oximetry System (ARPOS).

Current methods of measuring heart rate (HR) and oxygen levels (SPO2) require physical contact, are individualised, and for accurate oxygen levels may also require a blood test. No-touch or non-invasive technologies are not currently commercially available for use in healthcare settings. To date, there has been no assessment of a system that measures HR and SPO2 using commercial off-the-shelf camera technology that utilises R, G, B, and IR data. Moreover, no formal remote photoplethysmography studies have been performed in real-life scenarios with participants at home with different demographic characteristics. This novel study addresses all these objectives by developing, optimising, and evaluating a system that measures the HR and SPO2 of 40 participants. HR and SPO2 are determined by measuring the frequencies from different wavelength band regions using FFT and radiometric measurements after pre-processing face regions of interest (forehead, lips, and cheeks) from colour, IR, and depth data. Detrending, interpolating, hamming, and normalising the signal with FastICA produced the lowest RMSE of 7.8 for HR with the r-correlation value of 0.85 and RMSE 2.3 for SPO2. This novel system could be used in several critical care settings, including in care homes and in hospitals and prompt clinical intervention as required.

Ethics and acceptance of smart homes for older adults.

Societal challenges associated with caring for the physical and mental health of older adults worldwide have grown at an unprecedented pace, increasing demand for health-care services and technologies Despite the development of several assistive systems tailored to older adults, the rate of adoption of health technologies is low. This review discusses the ethical and acceptability challenges resulting in low adoption of health technologies specifically focused on smart homes for older adults. The findings have been structured in two categories: Ethical Considerations (Privacy, Social Support, and Autonomy) and Technology Aspects (User Context, Usability, and Training). The findings conclude that older adults community is more likely to adopt assistive systems when four key criteria are met. The technology should: be personalized toward their needs, protect their dignity and independence, provide user control, and not be isolating. Finally, we recommend researchers and developers working on assistive systems to: (1) provide interfaces via smart devices to control and configure the monitoring system with feedback for the user, (2) include various sensors/devices to architect a smart home solution in a way that is easy to integrate in daily life, and (3) define policies about data ownership.

Reproducibility of deep learning in digital pathology whole slide image analysis.

For a method to be widely adopted in medical research or clinical practice, it needs to be reproducible so that clinicians and regulators can have confidence in its use. Machine learning and deep learning have a particular set of challenges around reproducibility. Small differences in the settings or the data used for training a model can lead to large differences in the outcomes of experiments. In this work, three top-performing algorithms from the Camelyon grand challenges are reproduced using only information presented in the associated papers and the results are then compared to those reported. Seemingly minor details were found to be critical to performance and yet their importance is difficult to appreciate until the actual reproduction is attempted. We observed that authors generally describe the key technical aspects of their models well but fail to maintain the same reporting standards when it comes to data preprocessing which is essential to reproducibility. As an important contribution of the present study and its findings, we introduce a reproducibility checklist that tabulates information that needs to be reported in histopathology ML-based work in order to make it reproducible.

Generative Deep Learning in Digital Pathology Workflows.

Many modern histopathology laboratories are in the process of digitizing their workflows. Digitization of tissue images has made it feasible to research the augmentation or automation of clinical reporting and diagnosis. The application of modern computer vision techniques, based on deep learning, promises systems that can identify pathologies in slide images with a high degree of accuracy. Generative modeling is an approach to machine learning and deep learning that can be used to transform and generate data. It can be applied to a broad range of tasks within digital pathology, including the removal of color and intensity artifacts, the adaption of images in one domain into those of another, and the generation of synthetic digital tissue samples. This review provides an introduction to the topic, considers these applications, and discusses future directions for generative models within histopathology.

Gold nanorod reshaping in vitro and in vivo using a continuous wave laser.

Gold nanorods (GNRs) are increasingly being investigated for cancer theranostics as they possess features which lend themselves in equal measures as contrast agents and catalysts for photothermal therapy. Their optical absorption spectral peak wavelength is determined by their size and shape. Photothermal therapy using GNRs is typically established using near infrared light as this allows sufficient penetration into the tumour matrix. Continuous wave (CW) lasers are the most commonly applied source of near infrared irradiation on GNRs for tumour photothermal therapy. It is perceived that large tumours may require fractionated or prolonged irradiation. However the true efficacy of repeated or protracted CW irradiation on tumour sites using the original sample of GNRs remains unclear. In this study spectroscopy and transmission electron microscopy are used to demonstrate that GNRs reshape both in vitro and in vivo after CW irradiation, which reduces their absorption efficiency. These changes were sustained throughout and beyond the initial period of irradiation, resulting from a spectral blue-shift and a considerable diminution in the absorption peak of GNRs. Solid subcutaneous tumours in immunodeficient BALB/c mice were subjected to GNRs and analysed with electron microscopy pre- and post-CW laser irradiation. This phenomenon of thermally induced GNR reshaping can occur at relatively low bulk temperatures, well below the bulk melting point of gold. Photoacoustic monitoring of GNR reshaping is also evaluated as a potential clinical aid to determine GNR absorption and reshaping during photothermal therapy. Aggregation of particles was coincidentally observed following CW irradiation, which would further diminish the subsequent optical absorption capacity of irradiated GNRs. It is thus established that sequential or prolonged applications of CW laser will not confer any additional photothermal effect on tumours due to significant attenuations in the peak optical absorption properties of GNRs following primary laser irradiation.

Application of Gold Nanorods for Photothermal Therapy in Ex Vivo Human Oesophagogastric Adenocarcinoma.

Gold nanoparticles are chemically fabricated and tuned to strongly absorb near infrared (NIR) light, enabling deep optical penetration and therapy within human tissues, where sufficient heating induces tumour necrosis. In our studies we aim to establish the optimal gold nanorod (GNR) concentration and laser power for inducing hyperthermic effects in tissues and test this photothermal effect on ex vivo human oesophagogastric adenocarcinoma. The ideal GNR concentration and NIR laser power that would elicit sufficient hyperthermia for tumour necrosis was pre-determined on porcine oesophageal tissues. Human ex vivo oesophageal and gastric adenocarcinoma tissues were incubated with GNR solutions and a GNR-free control solution with corresponding healthy tissues for comparison, then irradiated with NIR light for 10 minutes. Temperature rise was found to vary linearly with both the concentration of GNRs and the laser power. Human ex vivo oesophageal and gastric tissues consistently demonstrated a significant temperature rise when incubated in an optimally concentrated GNR solution (3 x 10(10) GNRs/ml) prior to NIR irradiation delivered at an optimal power (2 W/cm2). A mean temperature rise of 27 degrees C was observed in tissues incubated with GNRs, whereas only a modest 2 degrees C rise in tissues not exposed to any GNRs. This study evaluates the photothermal effects of GNRs on oesophagogastric tissue examines their application in the minimally invasive therapeutics of oesophageal and gastric adenocarcinomas. This could potentially be an effective method of clinically inducing irreversible oesophagogastric tumour photodestruction, with minimal collateral damage expected in (healthy) tissues free from GNRs.

Aptamer-conjugated, fluorescent gold nanorods as potential cancer theradiagnostic agents.

GNRs are emerging as a new class of probes for theradiagnostic applications thanks to their unique optical properties. However, the achievement of proper nanoconstructs requires the synthesis of highly pure GNRs with well-defined aspect ratio (AR), in addition to extensive surface chemistry modification to provide them with active targeting and, possibly, multifunctionality. In this work, we refined the method of the seed mediated growth and developed a robust procedure for the fabrication of GNRs with specific AR. We also revealed and characterized unexplored aging phenomena that follow the synthesis and consistently alter GNRs' final AR. Such advances appreciably improved the feasibility of GNRs fabrication and offered useful insights on the growth mechanism. We next produced fluorescent, biocompatible, aptamer-conjugated GNRs by performing ligand exchange followed by bioconjugation to anti-cancer oligonucleotide AS1411. In vitro studies showed that our nanoconstructs selectively target cancer cells while showing negligible cytotoxicity. As a result, our aptamer-conjugated GNRs constitute ideal cancer-selective multifunctional probes and promising candidates as photothermal therapy agents.

Application of gold nanoparticles for gastrointestinal cancer theranostics: A systematic review.

Gold nanoparticles (GNPs) are readily synthesised structures that absorb light strongly to generate thermal energy which induces photothermal destruction of malignant tissue. This review examines the efficacy, potential challenges and toxicity from in vitro and in vivo applications of GNPs in oesophageal, gastric and colon cancers. A systematic literature search of Medline, Embase, Web of Science and Cochrane databases was performed using PRISMA guidelines. Two hundred and eighty-four papers were reviewed with sixteen studies meeting the inclusion criteria. The application of GNPs in eleven in vivo rodent studies with GI adenocarcinoma demonstrated excellent therapeutic outcomes but poor corroboration in terms of the cancer cells used, photothermal irradiation regimes, fluorophores and types of nanoparticles. There is compelling evidence of the translational potential of GNPs to be complimentary to surgery and feasible in the photothermal therapy of GI cancer but reproducibility and standardisation require development prior to GI cancer clinical trials. FROM THE CLINICAL EDITOR: Gold nanoparticles are one of the most potentially useful nanoparticles. This is especially true in cancer therapeutics because of their photothermal properties. In this comprehensive article, the authors reviewed the application and efficacy of gold nanoparticles in both the diagnosis and treatment of GI cancers. This review should provide a stimulus for researchers to further develop and translate these nanoparticles into future clinical trials.

Deformation-compensated averaging for clutter reduction in epiphotoacoustic imaging in vivo.

Photoacoustic imaging, based on ultrasound detected after laser irradiation, is an extension to diagnostic ultrasound for imaging the vasculature, blood oxygenation and the uptake of optical contrast media with promise for cancer diagnosis. For versatile scanning, the irradiation optics is preferably combined with the acoustic probe in an epi-style arrangement avoiding acoustically dense tissue in the acoustic propagation path from tissue irradiation to acoustic detection. Unfortunately epiphotoacoustic imaging suffers from strong clutter, arising from optical absorption in tissue outside the image plane, and from acoustic backscattering. This limits the imaging depth for useful photoacoustic image contrast to typically less than one centimeter. Deformation-compensated averaging (DCA), which takes advantage of clutter decorrelation induced by palpating the tissue with the imaging probe, has previously been proposed for clutter reduction. We demonstrate for the first time that DCA results in reduced clutter in real-time freehand clinical epiphotoacoustic imaging. For this purpose, combined photoacoustic and pulse-echo imaging at 10-Hz frame rate was implemented on a commercial scanner, allowing for ultrasound-based motion tracking inherently coregistered with photoacoustic frames. Results from the forearm and the neck confirm that contrast is improved and imaging depth increased by DCA.