Image Management System Research Articles

Validation of a whole slide image management system for metabolic-associated steatohepatitis for clinical trials.

The gold standard for enrollment and endpoint assessment in metabolic dysfunction-associated steatosis clinical trials is histologic assessment of a liver biopsy performed on glass slides. However, obtaining the evaluations from several expert pathologists on glass is challenging, as shipping the slides around the country or around the world is time-consuming and comes with the hazards of slide breakage. This study demonstrated that pathologic assessment of disease activity in steatohepatitis, performed using digital images on the AISight whole slide image management system, yields results that are comparable to those obtained using glass slides. The accuracy of scoring for steatohepatitis (nonalcoholic fatty liver disease activity score ≥4 with ≥1 for each feature and absence of atypical features suggestive of other liver disease) performed on the system was evaluated against scoring conducted on glass slides. Both methods were assessed for overall percent agreement with a consensus "ground truth" score (defined as the median score of a panel of three pathologists' glass slides). Each case was also read by three different pathologists, once on glass and once digitally with a minimum 2-week washout period between the modalities. It was demonstrated that the average agreement across three pathologists of digital scoring with ground truth was noninferior to the average agreement of glass scoring with ground truth [noninferiority margin: -0.05; difference: -0.001; 95% CI: (-0.027, 0.026); and p < 0.0001]. For each pathologist, there was a similar average agreement of digital and glass reads with glass ground truth (pathologist A, 0.843 and 0.849; pathologist B, 0.633 and 0.605; and pathologist C, 0.755 and 0.780). Here, we demonstrate that the accuracy of digital reads for steatohepatitis using digital images is equivalent to glass reads in the context of a clinical trial for scoring using the Clinical Research Network scoring system.

Artificial intelligence algorithm accurately assesses oestrogen receptor immunohistochemistry in metastatic breast cancer cytology specimens: A pilot study.

The Visiopharm artificial intelligence (AI) algorithm for oestrogen receptor (ER) immunohistochemistry (IHC) in whole slide images (WSIs) has been successfully validated in surgical pathology. This study aimed to assess its efficacy in cytology specimens. The study cohort comprised 105 consecutive cytology specimens with metastatic breast carcinoma. ER IHC WSIs were seamlessly integrated into the Visiopharm platform from the Image Management System (IMS) during our routine digital workflow, and an AI algorithm was employed for analysis. ER AI scores were compared with pathologists' manual consensus scores. Optimization steps were implemented and evaluated to reduce discordance. The overall concordance between pathologists' scores and AI scores was excellent (99/105, 94.3%). Six cases exhibited discordant results, including two false-negative (FN) cases due to abundant histiocytes incorrectly counted as negatively stained tumour cells by AI, two FN cases owing to weak staining, and two false-positive (FP) cases where pigmented macrophages were erroneously counted as positively stained tumour cells by AI. The Pearson correlation coefficient of ER-positive percentages between pathologists' and AI scores was 0.8483. Optimization steps, such as lowering the cut-off threshold and additional training using higher input magnification, significantly improved accuracy. The automated ER AI algorithm demonstrated excellent concordance with pathologists' assessments and accurately differentiated ER-positive from ER-negative metastatic breast carcinoma cytology cases. However, precision in identifying tumour cells in cytology specimens requires further enhancement.

Abstract 6178: Automating tumor purity assessment: Advanced AI-driven segmentation and classification of nuclei in H&amp;E stained slides

Abstract Background Tumor purity, the proportion of cancerous cells within a tumor mass, is a pivotal factor in oncological research and clinical practice and plays a significant role in the molecular testing workflow, particularly when selecting samples for genomic analysis. Accurate assessment of tumor purity is crucial for understanding tumor behavior, guiding treatment decisions, and predicting patient outcomes. Traditional methods of analyzing hematoxylin and eosin (H&amp;E) stained slides for tumor purity are predominantly pathologist manual readouts. This process is both time-consuming and prone to inter- and intra-observer variability, leading to inconsistencies in diagnosis and treatment planning. Here, we introduce an automated AI-based product designed to streamline tumor purity analysis. The RevealAI-HE digital assay leverages state-of-the-art AI algorithms to automate the segmentation and classification of nuclei in H&amp;E slides, enhancing the accuracy, consistency, and speed of tumor purity assessment. RevealAI-HE aims to assist pathologists in decision-making, support oncology research, streamline molecular testing workflows, and improve personalized care. Methods H&amp;E stained solid tumor slides were imaged using 3D-Histech Pannoramic 1000 and 250 scanners. From this dataset, 104 regions of interest (ROIs) were manually selected to cover a wide range of H&amp;E stain intensities in 14 different indications. A nuclear segmentation model based on StarDist-H&amp;E was deployed on these ROIs, and a pathology expert manually assigned one of three classifications to each nuclei: tumor, immune, and stroma. The resulting ground-truth dataset of over 26,000 classified nuclei was used for training a classifier model based on a ResNet architecture. For model evaluation, the nuclei dataset was randomly split into training (80%) and testing (20%) the model, and the training data was batched and randomized for each training epoch to ensure robustness and generalizability. The nuclear segmentation, nucleus classifier, and tumor purity statistics models were integrated into an end-to-end digital assay (RevealAI-HE) using Reveal Biosciences’ proprietary digital assay development pipeline. RevealAI-HE is integrated into Reveal Biosciences’ flagship whole slide image management system, ImageDx. Results The RevealAI-HE digital assay’s nucleus classification model achieved a high accuracy across all 14 indications. Compared to manual methods, the AI system significantly reduced the time required for tumor purity analysis. Furthermore, the AI model exhibited excellent consistency in its analyses, addressing the issue of variability seen in manual assessments. These results indicate that our AI product, in many aspects, surpasses the current gold standard of tumor purity analysis (pathologist readout), offering a more efficient, accurate, and reproducible approach. Citation Format: Nikolaus R. Wagner, Kaito Kikuchi, Kristen Fukuda, Nikola Spasic, Stacy Littlechild, Sinisa Todorovic, Dragana Stojnev, Masa Jovanovic, Claire Weston. Automating tumor purity assessment: Advanced AI-driven segmentation and classification of nuclei in H&amp;E stained slides [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6178.

Complete digital pathology transition: A large multi-center experience

IntroductionTransitioning from glass slide pathology to digital pathology for primary diagnostics requires an appropriate laboratory information system, an image management system, and slide scanners; it also reinforces the need for sophisticated pathology informatics including synoptic reporting. Previous reports have discussed the transition itself and relevant considerations for it, but not the selection criteria and considerations for the infrastructure. ObjectiveTo describe the process used to evaluate slide scanners, image management systems, and synoptic reporting systems for a large multisite institution. MethodsSix network hospitals evaluated six slide scanners, three image management systems, and three synoptic reporting systems. Scanners were evaluated based on the quality of image, speed, ease of operation, and special capabilities (including z-stacking, fluorescence and others). Image management and synoptic reporting systems were evaluated for their ease of use and capacity. ResultsAmong the scanners evaluated, the Leica GT450 produced the highest quality images, while the 3DHistech Pannoramic provided fluorescence and superior z-stacking. The newest generation of scanners, released relatively recently, performed better than slightly older scanners from major manufacturers Although the Olympus VS200 was not fully vetted due to not meeting all inclusion criteria, it is discussed herein due to its exceptional versatility. For Image Management Software, the authors believe that Sectra is, at the time of writing the best developed option, but this could change in the very near future as other systems improve their capabilities. All synoptic reporting systems performed impressively. ConclusionsSpecifics regarding quality and abilities of different components will change rapidly with time, but large pathology practices considering such a transition should be aware of the issues discussed and evaluate the most current generation to arrive at appropriate conclusions.

Requirements specification, design, and evaluation of dental image exchange and management system with user-centered approach: A case study in Iran.

Systems existing in hospital or clinic settings offer services within the physical environment. Examples of such systems include picture archiving and communication systems, which provide remote services for patients. To develop a successful system, methods like software development life cycles (SDLCs) and design techniques, such as prototyping, are needed. This study aimed to specify requirements, design, and evaluation of dental image exchange and management system using a user-centered approach. This cross-sectional study was conducted in four phases, each corresponding to different stages of SDLCs. User-needs data were used to gathered by interviews and observations. A prototype was developed using object-oriented programming and presented to users for feedback. Finally, focus group was used to finalized the prototype into the desired system. User needs were identified and prioritized from the outset, with ease of use, security, and mobile apps being their most essential requirements. The prototype underwent several iterations of design and evaluation in focus group sessions until users were satisfied, and their feedback was incorporated. Eventually, the prototype was refined into the final system with users' consent. The study revealed that instant access to information, voluntary participation, user interface (UI) design, and usefulness were critical variables for users and should be integral to any system. Successful implementation of such a system requires careful consideration of end-users' needs and their application to the system. Moreover, integrating the system with electronic health records can further enhance the treatment process and the efficiency of medical staff. The voluntary perspective of users played a significant role in achieving an exemplary UI and overall satisfaction with the system. Developers and policymakers should consider these aspects in similar system development projects.

Building a nonclinical pathology laboratory of the future for pharmaceutical research excellence

We describe a roadmap for a fully digital artificial intelligence (AI)-augmented nonclinical pathology laboratory across three continents. Underpinning the design are Good Laboratory Practice (GLP)-validated laboratory information management systems (LIMS), whole slide-scanners (WSS), image management systems (IMS), and a digital microscope intended for use by the nonclinical pathologist. Digital diagnostics are supported by tools that include AI-based virtual staining and deep learning-based decision support. Implemented during the COVID-19 pandemic, the initial digitized workflow largely mitigated disruption of pivotal nonclinical studies required to support pharmaceutical clinical testing. We believe that this digital transformation of our nonclinical pathology laboratories will promote efficiency and innovation in the future and enhance the quality and speed of drug development decision making.

BT07 Creation of a skin lesion image pipeline: key steps in developing a mobile application with DICOM® fields and metadata

Abstract A key issue facing artificial intelligence (AI) for dermatology is the lack of standardization around capture, transfer and storage of skin images and clinical data. To assist systematic capture of clinical data, we developed a nationally agreed proposed minimum dataset, based on Digital Imaging and Communications in Medicine (DICOM®) standards. Subsequently, we collaborated with an established image management system (IMS) supplier, Waba [supported by grants from our local National Health Service (NHS) innovation department], to incorporate these fields into a dermatology-specific mobile application (app), facilitating real-time capture and data transfer. We describe our steps in developing this app. The app integrates with the national Patient Administration System (SCI Store), allowing retrieval of real-time patient demographics linked to hospital or Community Health Index number. Each care episode captures a standardized dataset, including lesion views, clinical information and consent level. On episode completion, datasets are securely transferred to the IMS via WiFi or GPRS signal, following a fully auditable pathway. The IMS is a web-based app developed on a Microsoft Structured Query Language server, which allows authenticated users to access our trust’s centralized repository of nonradiological images. Adherence to industry-standard data storage format, patient safety reporting and specific access rights ensure compliance with Caldicott recommendations and data protection standards. For AI research, we are exploring conversion of datasets into DICOM format. Close collaboration between software developers and end users, including clinical photography, dermatology and general practice, was essential for app design. We worked on design architecture through fortnightly technical meetings, scheduled to progress development in months 1–8. Installation and functionality testing was performed in months 9 and 10, followed by comprehensive user testing in months 10–12. Live app test-of-change rollout was performed in months 13–16, with continued usability refinement during this period. To develop a standardized minimum dataset, we conducted a consensus exercise with the British Association of Dermatology AI working group. Based on DICOM standards, a comprehensive international standard for medical imaging, this could improve the quality, availability and interoperability of data for AI development. In summary, we have developed an app to standardize the capture, transmission and storage of skin lesion images and associated metadata. By collaborating with an industry partner with NHS experience, iterative user-led design and national engagement for standard-setting, we believe this is a useful step toward a national UK skin database for training and validating AI systems. Future work is planned to evaluate the output of this data pipeline.

BT05 Community and Locality Imaging Centres: innovation in the skin cancer care pathway

Abstract Timely management of patients referred with suspected skin cancer remains challenging due to demand that exceeds the capacity of dermatology departments. Encouraging the attachment of images and key metadata along with referrals from primary care helps promote enhanced clinical referral triage. We describe the process of establishing Community and Locality Imaging Centres (CLICs) that seek not only to promote efficient triage, but also to establish a national pipeline of anonymized images, with linked Digital Imaging and Communications in Medicine (DICOM®) data, to assess the potential for future use of artificial intelligence (AI) algorithms in clinical practice. General practitioners (GPs) can request a CLIC appointment at the time of referral, removing the time pressure and potential technical barriers associated with photography and image attachment. Each centre follows the same standardized image capture and hardware operating procedure, ensuring that high-quality, in-focus digital images are consistently acquired. The CLIC appointments take an average of 15 min, during which a suitably trained healthcare professional obtains written patient consent and, using a National Health Service-approved mobile phone with a dermatoscope attachment, takes paired macroscopic and dermoscopic views of each lesion. Relevant metadata, in line with the recently agreed DICOM standards for dermatology, are recorded for individual lesions, enhancing the utility of each photograph. Paired images and metadata are regularly archived to a secure medical image management system and are linked with the referral, allowing dermatologists to perform enhanced clinical triage and communicate one of the following outcomes directly to the referring GP: routine dermatology appointment, urgent dermatology appointment, direct to biopsy or benign lesion–reassurance and discharge. This process aims to streamline the primary care referral pathway and promote immediate benefits to patient care by reducing unnecessary travel for face-to-face outpatient appointments. The establishment of CLICs in Scotland supports referral optimization by providing an efficient, convenient option for GPs and patients to obtain high-quality clinical photography. This model allows for flexible integration into local services: hospital-based with medical photographers, intermediate secondary care-led centres or community-based primary care hubs. Suspected skin cancer referrals can be efficiently triaged by dermatologists when paired with macroscopic and dermoscopic images and relevant patient metadata compliant with DICOM standards. The same images and metadata, once anonymized, can be securely transferred to a nationwide database for deep learning AI algorithm development. Validated AI programs, with the aim of being deployed at the primary care level, inform opportunities for further enhancement of the referral pathway.

Efficient Management and Scheduling of Massive Remote Sensing Image Datasets

The rapid development of remote sensing image sensor technology has led to exponential increases in available image data. The real-time scheduling of gigabyte-level images and the storage and management of massive image datasets are incredibly challenging for current hardware, networking and storage systems. This paper’s three novel strategies (ring caching, multi-threading and tile-prefetching mechanisms) are designed to comprehensively optimize the remote sensing image scheduling process from image retrieval, transmission and visualization perspectives. A novel remote sensing image management and scheduling system (RSIMSS) is designed using these three strategies as its core algorithm, the PostgreSQL database and HDFS distributed file system as its underlying storage system, and the multilayer Hilbert spatial index and image tile pyramid to organize massive remote sensing image datasets. Test results show that the RSIMSS provides efficient and stable image storage performance and allows real-time image scheduling and view roaming.

A whole-slide imaging based workflow reduces the reading time of pathologists.

Even though entirely digitized microscopic tissue sections (whole slide images, WSIs) are increasingly being used in histopathology diagnostics, little data is still available on the effect of this technique on pathologists' reading time. This study aimed to compare the time required to perform the microscopic assessment by pathologists between a conventional workflow (an optical microscope) and digitized WSIs. WSI was used in primary diagnostics at the Laboratory for Pathology Eastern Netherlands for several years (LabPON, Hengelo, The Netherlands). Cases were read either in a traditional workflow, with the pathologist recording the time required for diagnostics and reporting, or entirely digitally. Reading times were extracted from image management system log files, and the digitized workflow was fully integrated into the laboratory information system. The digital workflow saved time in the majority of case categories, with prostate biopsies saving the most (68% time gain). Taking into account case distribution, the digital workflow produced an average gain of 12.3%. Using WSI instead of conventional microscopy significantly reduces pathologists' reading times. Pathologists must work in a fully integrated environment to fully reap the benefits of a digital workflow.

The Pandemic as a Catalyst for Digital Pathology Adoption as an LDT

Abstract Whole-slide images (WSI) are the basis for the application of artificial intelligence/machine learning and other informatics methods to histological diagnosis and will further blur the line separating anatomic and clinical pathology. FDA classified WSI systems for primary diagnosis as class III (highest risk) medical devices until 2017. This discouraged anatomic pathology laboratories at risk-averse domestic institutions like mine from investing in these digital pathology (DP) platforms. In 2017, FDA downgraded WSI to class II (moderate risk) when they de-novo approved a system marketed by Philips. We were not interested in that system at my institution, but the downgrade caused us to reset our perception of the risk of validating a RUO system for primary diagnosis. Cost remained a barrier. In April 2020, FDA issued temporary guidance stating they would not enforce premarket approval of WSI systems to facilitate pathologists working remotely during the SARS-CoV-2 pandemic. The guidance included a statement that “laboratories and hospitals consider performing a validation study.” In January 2021, FDA proposed making the temporary non-enforcement guidance permanent. So, in a little more than three years, WSI for primary diagnosis had gone from class III to exempted from pre-market approval! This nicely aligned the approval framework for WSI with the approval framework for our conventional optical microscopes, which are statutorily exempted from approval, and further reset our perception of risk. In April 2021, FDA withdrew the proposal to make non-enforcement permanent, but the temporary non-enforcement guidance is still in effect at the time of writing. Amid all this FDA activity, the College of American Pathologists updated and reissued their consensus guidelines for validating WSI systems for diagnostic purposes in March 2021. The narrative mentions the FDA’s recent approval of a few WSI systems and anticipates more, but the expert panel recommendations do not include any related to the approval status of systems. The reissue of this document reminded us that, as clinical laboratorians, we are capable of safely validating WSI as a laboratory-developed test and are supported in doing so by consensus guidelines from one of our leading professional organizations. In early 2021 we committed to funding a DP initiative to make WSI part of our routine histological process for 10% of our anatomic pathology cases. The initial capital investment is $1.5M. When realized, the microscope slides for designated pathology services will be transported directly from the cover slipper to a slide scanner and electronically distributed to pathologists using a clinical-grade image management system that we share with our radiology department. We made the decision to fund this in the context of the regulatory (decreased perception of risk), sociological (demand for remote telepathology), and technological (availability of scalable WSI systems) changes that occurred during the pandemic.

Decentralized Patient-Centric Report and Medical Image Management System Based on Blockchain Technology and the Inter-Planetary File System.

Several academicians have been actively contributing to establishing a practical solution to storing and distributing medical images and test reports in the research domain of health care in recent years. Current procedures mainly rely on cloud-assisted centralized data centers, which raise maintenance expenditure, necessitate a large amount of storage space, and raise privacy concerns when exchanging data across a network. As a result, it is critically essential to provide a framework that allows for the efficient exchange and storage of large amounts of medical data in a secure setting. In this research, we describe a unique proof-of-concept architecture for a distributed patient-centric test report and image management (PCRIM) system that aims to facilitate patient privacy and control without the need for a centralized infrastructure. We used an Ethereum blockchain and a distributed file system technology called the Inter-Planetary File System in this system (IPFS). Then, to secure a distributed and trustworthy access control policy, we designed an Ethereum smart contract termed the patient-centric access control protocol. The IPFS allows for the decentralized storage of medical metadata, such as images, with worldwide accessibility. We demonstrate how the PCRIM system design enables hospitals, patients, and image requestors to obtain patient-centric data in a distributed and secure manner. Finally, we tested the proposed framework in the Windows environment by deploying a smart contract prototype on an Ethereum TESTNET blockchain. The findings of the study indicate that the proposed strategy is both efficient and practicable.

4SA: OPTIMIZING SPACE FILLING CURVE BASED GRID CELL INDEXING TO SCALABLY MANAGE REMOTELY SENSED IMAGES IN KEY-VALUE DATABASES

Abstract. State-of-the-art remote sensing image management systems adopt scalable databases and employ sophisticated indexing techniques to perform window and containment queries. Many rely on space-filling curve (SFC) based index techniques designed for key-value databases and are predominantly employable for images that are iso-oriented. Critically, these indexes do not consider the high degree of overlap among images that exists in many data sets and the affiliated storage requirements. Specifically, employing an SFC-based grid cell index approach in consort with ground footprint coverage of the images requires storage of a unique image object identification (IOI) for each image in every grid cell where overlap occurs. Such an approach adversely affects both storage and query response times. In response, this paper presents an optimization technique for an SFC-based grid cell space indexing. The optimization is specifically designed for window and containment queries where the region of interest overlaps with at least a 2 × 2 grid of cells. The technique is based on four cell removal steps, thus called “four step algorithm” (4SA). Each step employs a unique spatial configuration to check for continuous spatial extent. If present, the IOI of the target cell is omitted from further consideration. Analysis and experiments on real world and synthetic image data demonstrated that 4SA improved storage demands by 41.3% – 47.8%. Furthermore, in the performed querying experiments, only 42% of IOI elements needed to be processed, thus yielding a 58% productivity gain. The reduction of IOI elements in querying also impacted the CPU execution time (3.0% – 5.2%). The 4SA also demonstrated data scalability and concurrent user scalability in querying large regions by completing the index searching and concurrent user scalability 1.86% – 3.35% faster than when 4SA was not applied.

Pigeon Inspired Optimization with Encryption Based Secure Medical Image Management System.

Presently, technological advancements in the healthcare sector pose a challenging problem relevant to the security and privacy of health-related applications. Medical images can be considered significant and sensitive data in the medical informatics system. In order to transmit medical images in an open medium, the design of secure encryption algorithms becomes essential. Encryption can be considered one of the effective solutions for accomplishing security. Although numerous models have existed in the literature, they could not adaptable to the rising number of medicinal images in the health sector. At the same time, the optimal key generation process acts as a vital part in defining the performance of the encryption techniques. Therefore, this article presents a Pigeon Inspired Optimization with Encryption-based Secure Medical Image Management (PIOE-SMIM) technique. The proposed PIOE-SMIM approach majorly concentrates on the development of secret share creation (SSC) and the encryption process. At the initial stage, the medical images are converted into a collection of 12 shares using the SSC approach. In addition, an elliptic curve cryptography (ECC) scheme is employed for the encryption process. In order to optimum key creation procedure in the ECC model, the PIO technique is exploited with the aim of maximizing PSNR. Finally, on the receiver side, the decryption and share reconstruction processes are performed to construct the original images. The PIOE-SMIM model displayed an enhanced PSNR of 59.37 dB in image 1. Improved PSNR of 59.53 dB is given for image 5 using the PIOE-SMIM model. For demonstrating an enhanced performance of the PIOE-SMIM method, a widespread experimental study is made and the results highlighted the supremacy of the PIOE-SMIM model over other techniques.

A data management infrastructure for the integration of imaging and omics data in life sciences

BackgroundAs technical developments in omics and biomedical imaging increase the throughput of data generation in life sciences, the need for information systems capable of managing heterogeneous digital assets is increasing. In particular, systems supporting the findability, accessibility, interoperability, and reusability (FAIR) principles of scientific data management.ResultsWe propose a Service Oriented Architecture approach for integrated management and analysis of multi-omics and biomedical imaging data. Our architecture introduces an image management system into a FAIR-supporting, web-based platform for omics data management. Interoperable metadata models and middleware components implement the required data management operations. The resulting architecture allows for FAIR management of omics and imaging data, facilitating metadata queries from software applications. The applicability of the proposed architecture is demonstrated using two technical proofs of concept and a use case, aimed at molecular plant biology and clinical liver cancer research, which integrate various imaging and omics modalities.ConclusionsWe describe a data management architecture for integrated, FAIR-supporting management of omics and biomedical imaging data, and exemplify its applicability for basic biology research and clinical studies. We anticipate that FAIR data management systems for multi-modal data repositories will play a pivotal role in data-driven research, including studies which leverage advanced machine learning methods, as the joint analysis of omics and imaging data, in conjunction with phenotypic metadata, becomes not only desirable but necessary to derive novel insights into biological processes.

Integrating and validating automated digital imaging analysis of estrogen receptor immunohistochemistry in a fully digital workflow for clinical use

BackgroundThe Visiopharm automated estrogen receptor (ER) digital imaging analysis (DIA) algorithm assesses digitized ER immunohistochemistry (IHC) by segmenting tumor nuclei and detecting stained nuclei automatically. We aimed to integrate and validate this algorithm in a digital pathology workflow for clinical use. DesignThe study cohort consisted of a serial collection of 97 invasive breast carcinoma specimens including 73 biopsies and 24 resections. ER IHC slides were scanned into Philips Image Management System (IMS) during our routine digital workflow and digital images were directly streamed into Visiopharm platform and analyzed using automated ER algorithm to obtain the positively stained tumor nuclei and staining intensity. ER DIA scores were compared with pathologists’ manual scores. ResultsThe overall concordance between pathologists’ reads and DIA reads was excellent (91/97, 93.8%). Pearson Correlation Coefficient of the percentage of ER positive nuclei between the original reads and VIS reads was 0.72. Six cases (3 ER-negative and 3 ER-positive) had discordant results. All 3 false negative cases had very weak ER staining and no more than 10% positivity. The causes for false positive DIA were mainly pre-analytic/pre-imaging and included intermixed benign glands in tumor area, ductal carcinoma in-situ (DCIS) components, and tissue folding. ConclusionsAutomated ER DIA demonstrates excellent concordance with pathologists’ scores and accurately discriminates ER positive from negative cases. Furthermore, integrating automated biomarker DIA into a busy clinical digital workflow is feasible and may save time and labor for pathologists.

Inpatient Teledermatology Referrals During the COVID-19 Pandemic in a UK Trust: A Comparative Review and Doctor Survey

Background The COVID-19 pandemic has broadened the scope of teledermatology services in the United Kingdom from a primarily outpatient-based triage tool to the management of inpatient referrals. In order to reduce the risk of transmission in hospital, a number of changes were implemented within our department. As part of this, our on-call referrals were transferred to a telemedicine app, which incorporates the secure transfer of user-generated patient images onto a web-based image management system providing remote access for the dermatology team. Objective This study aimed to compare how the introduction of this referral method impacted the nature and number of referrals received, the efficiency of the on-call service, and user preferences. Methods A retrospective cohort study was conducted to compare the number of referrals, time taken to review, and referral diagnoses between previous referral methods to the dermatology department (bleep, fax, email) (July and September 2019) and the new teledermatology app (July and September 2020). We also performed a survey of junior doctors, seeking their feedback and preferences pertaining to the new referral system. Results The number of referrals increased by 80%, with a 6-fold increase in lesion referrals. There is a possibility that not all referrals from 2019 were accounted for as paper documents are easily lost or discarded, highlighting another advantage of teledermatology in providing a reliable record of referrals. Dermatology referrals may have increased as the telemedicine app is more accessible to staff across sites. The telemedicine app also led to a reduction in time to review by 0.53 days, resulting in a significantly higher number of patients being given dermatology input on the day of the referral (78% vs 58%). This will have led to earlier treatment, improved patient outcomes, and shorter inpatient stays, resulting in potential cost reductions for the hospital. The survey of junior doctors showed that 81% preferred teledermatology to the previous referral methods. Conclusions The introduction of teledermatology has provided an effective and acceptable method of managing on-call dermatology referrals. Easier access to dermatology advice via teledermatology may result in higher numbers of referrals, which may warrant strict referral criteria to prevent oversubscription of the on-call service. Teledermatology ensures an accurate log of referrals, including the nature of referrals, allowing for better auditing and service improvement. Teledermatology referrals allow for advice to be provided within shorter time frames compared to previous methods. This should improve patient outcomes and reduce hospital admission stays, potentially resulting in cost savings for the hospital. Conflict of Interest None declared.

Deep neural models for automated multi-task diagnostic scan management—quality enhancement, view classification and report generation

The detailed physiological perspectives captured by medical imaging provides actionable insights to doctors to manage comprehensive care of patients. However, the quality of such diagnostic image modalities is often affected by mismanagement of the image capturing process by poorly trained technicians and older/poorly maintained imaging equipment. Further, a patient is often subjected to scanning at different orientations to capture the frontal, lateral and sagittal views of the affected areas. Due to the large volume of diagnostic scans performed at a modern hospital, adequate documentation of such additional perspectives is mostly overlooked, which is also an essential key element of quality diagnostic systems and predictive analytics systems. Another crucial challenge affecting effective medical image data management is that the diagnostic scans are essentially stored as unstructured data, lacking a well-defined processing methodology for enabling intelligent image data management for supporting applications like similar patient retrieval , automated disease prediction etc. One solution is to incorporate automated diagnostic image descriptions of the observation/findings by leveraging computer vision and natural language processing. In this work, we present multi-task neural models capable of addressing these critical challenges. We propose ESRGAN, an image enhancement technique for improving the quality and visualization of medical chest x-ray images, thereby substantially improving the potential for accurate diagnosis, automatic detection and region-of-interest segmentation. We also propose a CNN-based model called ViewNet for predicting the view orientation of the x-ray image and generating a medical report using Xception net, thus facilitating a robust medical image management system for intelligent diagnosis applications. Experimental results are demonstrated using standard metrics like BRISQUE, PIQE and BLEU scores, indicating that the proposed models achieved excellent performance. Further, the proposed deep learning approaches enable diagnosis in a lesser time and their hybrid architecture shows significant potential for supporting many intelligent diagnosis applications.

Blockchain Enabled Emperor Penguin Optimizer Based Encryption Technique for Secure Image Management System

In recent years, the electronic sharing of digital images faces a major threat to security, as the existing image transmission infrastructure is mainly based on the trust of third parties. At the same time, the available solutions are placed on the cloud based centralized data center, which is expensive, requires large storage area, and security issues regarding the transmission of data over the network. So, it is needed to develop an image management system which enables sharing and storing of digital images effectively. This paper develops novel multiple share creation schemes with block technology for secure image management (MSCCBT-SIM) systems. The MSCCBT-SIM model allows the user to create consensus with no dependencies on central authorities. It involves an MSC which involves share creation and share encryption using emperor penguin optimizer based ElGamal public key cryptosystem (EPO-EPKC). In addition, the blockchain is used as a distributed data storage mechanism to generate a ledger for permitting access to the user and prevent third party access to the encrypted shares. The application of blockchain technology and MSC techniques helps to achieve decentralization, highly reliable, inexpensive, and secure transmission and storage of digital images. In order to validate the effective performance of the MSCCBT-SIM model, a series of simulations take place and investigated the results interms of different measures. The experimental results ensured the better performance of the MSCCBT-SIM model with the superior PSNR of 57.09 dB whereas the HOCE-ECC, GO-ECC, PSO-ECC, and CS-ECC methods offered a lower PSNR of 53.08 dB, 52.91 dB, 52.65 dB, and 50.51 dB respectively.

Implementing an Integrated Large-Scale Clinical Information System for ISSSTE\u2019s Hospital Network in Mexico

The Institute for Security and Social Services for State Workers (ISSSTE) is a large public provider of health care services that serve around 13.2 million Mexican government workers and their families. To attain process efficiencies, cost reductions, and improvement of the quality of diagnostic and imaging services, ISSSTE was set out in 2019 to create a digital filmless medical image and report management system. A large-scale clinical information system (CIS), including radiology information system (RIS), picture archiving and communication system (PACS), and clinical data warehouse (CDW) components, was implemented at ISSSTE’s network of forty secondary- and tertiary-level public hospitals, applying global HL-7 and Digital Imaging and Communications in Medicine (DICOM) standards. In just 5 months, 40 hospitals had their endoscopy, radiology, and pathology services functionally interconnected within a national CIS and RIS/PACS on secure private local area networks (LANs) and a secure national wide area network (WAN). More than 2 million yearly studies and reports are now in digital form in a CDW, securely stored and always available. Benefits include increased productivity, reduced turnaround times, reduced need for duplicate exams, and reduced costs. Functional IT solutions allow ISSSTE hospitals to leave behind the use of radiographic film and printed medical reports with important cost reductions, as well as social and environmental impacts, leading to direct improvement in the quality of health care services rendered.