Discriminant analysis and machine learning approach for evaluating and improving the performance of immunohistochemical algorithms for COO classification of DLBCL

Role of artificial intelligence in haematolymphoid diagnostics

The advent of digital pathology and the deployment of high-throughput molecular techniques are generating an unprecedented mass of data. Thanks to advances in computational sciences, artificial intelligence (AI) approaches represent a promising avenue for extracting relevant information from complex data structures. From diagnostic assistance to powerful research tools, the potential fields of application of machine learning techniques in pathology are vast and constitute the subject of considerable research work. The aim of this article is to provide an overview of the potential applications of AI in the field of haematopathology and to define the role that these emerging technologies could play in our laboratories in the short to medium term.

Prognosis Prediction of Diffuse Large B-Cell Lymphoma in 18F-FDG PET Images Based on Multi-Deep-Learning Models

Diffuse large B-cell lymphoma (DLBCL), a cancer of B cells, has been one of the most challenging and complicated diseases because of its considerable variation in clinical behavior, response to therapy, and prognosis. Radiomic features from medical images, such as PET images, have become one of the most valuable features for disease classification or prognosis prediction using learning-based methods. In this paper, a new flexible ensemble deep learning model is proposed for the prognosis prediction of the DLBCL in 18F-FDG PET images. This study proposes the multi-R-signature construction through selected pre-trained deep learning models for predicting progression-free survival (PFS) and overall survival (OS). The proposed method is trained and validated on two datasets from different imaging centers. Through analyzing and comparing the results, the prediction models, including Age, Ann abor stage, Bulky disease, SUVmax, TMTV, and multi-R-signature, achieve the almost best PFS prediction performance (C-index: 0.770, 95% CI: 0.705-0.834, with feature adding fusion method and C-index: 0.764, 95% CI: 0.695-0.832, with feature concatenate fusion method) and OS prediction (C-index: 0.770 (0.692-0.848) and 0.771 (0.694-0.849)) on the validation dataset. The developed multiparametric model could achieve accurate survival risk stratification of DLBCL patients. The outcomes of this study will be helpful for the early identification of high-risk DLBCL patients with refractory relapses and for guiding individualized treatment strategies.

Cardiac "hypertrophy" phenotyping: differentiating aetiologies with increased left ventricular wall thickness on echocardiography

Aims

Differentiating phenotypes of cardiac "hypertrophy" characterised by increased wall thickness on echocardiography is essential for management and prognostication. Transthoracic echocardiography is the most commonly used screening test for this purpose. We sought to identify echocardiographic markers that distinguish infiltrative and storage disorders that present with increased left ventricular (LV) wall thickness, namely, cardiac amyloidosis (CA) and Anderson-Fabry disease (AFD), from hypertensive heart disease (HHT).

Methods

Patients were retrospectively recruited from Westmead Hospital, Sydney, and Princess Alexandra Hospital, Brisbane. LV structural, systolic, and diastolic function parameters, as well as global (LVGLS) and segmental longitudinal strains, were assessed. Previously reported echocardiographic parameters including relative apical sparing ratio (RAS), LV ejection fraction-to-strain ratio (EFSR), mass-to-strain ratio (MSR) and amyloidosis index (AMYLI) score (relative wall thickness × E/e') were evaluated.

Results

A total of 209 patients {120 CA [58 transthyretin amyloidosis (ATTR) and 62 light-chain (AL) amyloidosis], 31 AFD and 58 HHT patients; mean age 64.1 ± 13.7 years, 75% male} comprised the study cohort. Echocardiographic measurements differed across the three groups, The LV mass index was higher in both CA {median 126.6 [interquartile range (IQR) 106.4-157.9 g/m²]} and AFD [median 134 (IQR 108.8-152.2 g/m²)] vs. HHT [median 92.7 (IQR 79.6-102.3 g/m²), p < 0.05]. LVGLS was lowest in CA [median 12.29 (IQR 10.33-15.56%)] followed by AFD [median 16.92 (IQR 14.14-18.78%)] then HHT [median 18.56 (IQR 17.51-19.97%), p < 0.05]. Diastolic function measurements including average e' and E/e' were most impaired in CA and least impaired in AFD. Indexed left atrial volume was highest in CA. EFSR and MSR differentiated secondary (CA + AFD) from HHT [receiver operating curve-area under the curve (ROC-AUC) of 0.80 and 0.91, respectively]. RAS and AMYLI score differentiated CA from AFD (ROC-AUC of 0.79 and 0.80, respectively). A linear discriminant analysis with stepwise variable selection using linear combinations of LV mass index, average e', LVGLS and basal strain correctly classified 79% of all cases.

Conclusion

Simple echocardiographic parameters differentiate between different "hypertrophic" cardiac phenotypes. These have potential utility as a screening tool to guide further confirmatory testing.

An open source pipeline for quantitative immunohistochemistry image analysis of inflammatory skin disease using artificial intelligence

BACKGROUND

The application of artificial intelligence (AI) to whole slide images has the potential to improve research reliability and ultimately diagnostic efficiency and service capacity. Image annotation plays a key role in AI and digital pathology. However, the work-streams required for tissue-specific (skin) and immunostain-specific annotation has not been extensively studied compared with the development of AI algorithms.

OBJECTIVES

The objective of this study is to develop a common workflow for annotating whole slide images of biopsies from inflammatory skin disease immunostained with a variety of epidermal and dermal markers prior to the development of the AI-assisted analysis pipeline.

METHODS

A total of 45 slides containing 3-5 sections each were scanned using Aperio AT2 slide scanner (Leica Biosystems). These slides were annotated by hand using a commonly used image analysis tool which resulted in more than 4000 images blocks. We used deep learning (DL) methodology to first sequentially segment (epidermis and upper dermis), with the exclusion of common artefacts and second to quantify the immunostained signal in those two compartments of skin biopsies and the ratio of positive cells.

RESULTS

We validated two DL models using 10-fold validation runs and by comparing to ground truth manually annotated data. The models achieved an average (global) accuracy of 95.0% for the segmentation of epidermis and dermis and 86.1% for the segmentation of positive/negative cells.

CONCLUSIONS

The application of two DL models in sequence facilitates accurate segmentation of epidermal and dermal structures, exclusion of common artefacts and enables the quantitative analysis of the immunostained signal. However, inaccurate annotation of the slides for training the DL model can decrease the accuracy of the output. Our open source code will facilitate further external validation across different immunostaining platforms and slide scanners.

Prognostic molecular biomarkers in diffuse large B-cell lymphoma in the rituximab era and their therapeutic implications

Diffuse large B-cell lymphoma (DLBCL) represents a group of tumors characterized by substantial heterogeneity in terms of their pathological and biological features, a causal factor of their varied clinical outcome. This variation has persisted despite the implementation of rituximab in treatment regimens over the last 20 years. In this context, prognostic biomarkers are of great importance in order to identify high-risk patients that might benefit from treatment intensification or the introduction of novel therapeutic agents. Herein, we review current knowledge on specific immunohistochemical or genetic biomarkers that might be useful in clinical practice. Gene-expression profiling is a tool of special consideration in this effort, as it has enriched our understanding of DLBCL biology and has allowed for the classification of DLBCL by cell-of-origin as well as by more elaborate molecular signatures based on distinct gene-expression profiles. These subgroups might outperform individual biomarkers in terms of prognostication; however, their use in clinical practice is still limited. Moreover, the underappreciated role of the tumor microenvironment in DLBCL prognosis is discussed in terms of prognostic gene-expression signatures, as well as in terms of individual biomarkers of prognostic significance. Finally, the efficacy of novel therapeutic agents for the treatment of DLBCL patients are discussed and an evidence-based therapeutic approach by specific genetic subgroup is suggested.

Artificial Intelligence and Digital Microscopy Applications in Diagnostic Hematopathology

Digital Pathology is the process of converting histology glass slides to digital images using sophisticated computerized technology to facilitate acquisition, evaluation, storage, and portability of histologic information. By its nature, digitization of analog histology data renders it amenable to analysis using deep learning/artificial intelligence (DL/AI) techniques. The application of DL/AI to digital pathology data holds promise, even if the scope of use cases and regulatory framework for deploying such applications in the clinical environment remains in the early stages. Recent studies using whole-slide images and DL/AI to detect histologic abnormalities in general and cancer in particular have shown encouraging results. In this review, we focus on these emerging technologies intended for use in diagnostic hematology and the evaluation of lymphoproliferative diseases.

Molecular Complexity of Diffuse Large B-Cell Lymphoma: Can It Be a Roadmap for Precision Medicine?

Diffuse large B-cell lymphoma (DLBCL) is the most common non-Hodgkin lymphoma; it features extreme molecular heterogeneity regardless of the classical cell-of-origin (COO) classification. Despite this, the standard therapeutic approach is still immunochemotherapy (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone-R-CHOP), which allows a 60% overall survival (OS) rate, but up to 40% of patients experience relapse or refractory (R/R) disease. With the purpose of searching for new clinical parameters and biomarkers helping to make a better DLBCL patient characterization and stratification, in the last years a series of large discovery genomic and transcriptomic studies has been conducted, generating a wealth of information that needs to be put in order. We reviewed these researches, trying ultimately to understand if there are bases offering a roadmap toward personalized and precision medicine also for DLBCL.