Clinical-grade Computational Pathology: Alea Iacta Est

Histopathological Image Deep Feature Representation for CBIR in Smart PACS

Pathological Anatomy is moving toward computerizing processes mainly due to the extensive digitization of histology slides that resulted in the availability of many Whole Slide Images (WSIs). Their use is essential, especially in cancer diagnosis and research, and raises the pressing need for increasingly influential information archiving and retrieval systems. Picture Archiving and Communication Systems (PACSs) represent an actual possibility to archive and organize this growing amount of data. The design and implementation of a robust and accurate methodology for querying them in the pathology domain using a novel approach are mandatory. In particular, the Content-Based Image Retrieval (CBIR) methodology can be involved in the PACSs using a query-by-example task. In this context, one of many crucial points of CBIR concerns the representation of images as feature vectors, and the accuracy of retrieval mainly depends on feature extraction. Thus, our study explored different representations of WSI patches by features extracted from pre-trained Convolution Neural Networks (CNNs). In order to perform a helpful comparison, we evaluated features extracted from different layers of state-of-the-art CNNs using different dimensionality reduction techniques. Furthermore, we provided a qualitative analysis of obtained results. The evaluation showed encouraging results for our proposed framework.

A Survival Guide for the Rapid Transition to a Fully Digital Workflow: The "Caltagirone Example"

Digital pathology for the routine assessment of cases for primary diagnosis has been implemented by few laboratories worldwide. The Gravina Hospital in Caltagirone (Sicily, Italy), which collects cases from 7 different hospitals distributed in the Catania area, converted the entire workflow to digital starting from 2019. Before the transition, the Caltagirone pathology laboratory was characterized by a non-tracked workflow, based on paper requests, hand-written blocks and slides, as well as manual assembling and delivering of the cases and glass slides to the pathologists. Moreover, the arrangement of the spaces and offices in the department was illogical and under-productive for the linearity of the workflow. For these reasons, an adequate 2D barcode system for tracking purposes, the redistribution of the spaces inside the laboratory and the implementation of the whole-slide imaging (WSI) technology based on a laboratory information system (LIS)-centric approach were adopted as a needed prerequisite to switch to a digital workflow. The adoption of a dedicated connection for transfer of clinical and administrative data between different software and interfaces using an internationally recognised standard (Health Level 7, HL7) in the pathology department further facilitated the transition, helping in the integration of the LIS with WSI scanners. As per previous reports, the components and devices chosen for the pathologists' workstations did not significantly impact on the WSI-based reporting phase in primary histological diagnosis. An analysis of all the steps of this transition has been made retrospectively to provide a useful "handy" guide to lead the digital transition of "analog", non-tracked pathology laboratories following the experience of the Caltagirone pathology department. Following the step-by-step instructions, the implementation of a paperless routine with more standardized and safe processes, the possibility to manage the priority of the cases and to implement artificial intelligence (AI) tools are no more an utopia for every "analog" pathology department.

An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study

BACKGROUND

There is high demand to develop computer-assisted diagnostic tools to evaluate prostate core needle biopsies (CNBs), but little clinical validation and a lack of clinical deployment of such tools. We report here on a blinded clinical validation study and deployment of an artificial intelligence (AI)-based algorithm in a pathology laboratory for routine clinical use to aid prostate diagnosis.

METHODS

An AI-based algorithm was developed using haematoxylin and eosin (H&E)-stained slides of prostate CNBs digitised with a Philips scanner, which were divided into training (1 357 480 image patches from 549 H&E-stained slides) and internal test (2501 H&E-stained slides) datasets. The algorithm provided slide-level scores for probability of cancer, Gleason score 7-10 (vs Gleason score 6 or atypical small acinar proliferation [ASAP]), Gleason pattern 5, and perineural invasion and calculation of cancer percentage present in CNB material. The algorithm was subsequently validated on an external dataset of 100 consecutive cases (1627 H&E-stained slides) digitised on an Aperio AT2 scanner. In addition, the AI tool was implemented in a pathology laboratory within routine clinical workflow as a second read system to review all prostate CNBs. Algorithm performance was assessed with area under the receiver operating characteristic curve (AUC), specificity, and sensitivity, as well as Pearson's correlation coefficient (Pearson's r) for cancer percentage.

FINDINGS

The algorithm achieved an AUC of 0·997 (95% CI 0·995 to 0·998) for cancer detection in the internal test set and 0·991 (0·979 to 1·00) in the external validation set. The AUC for distinguishing between a low-grade (Gleason score 6 or ASAP) and high-grade (Gleason score 7-10) cancer diagnosis was 0·941 (0·905 to 0·977) and the AUC for detecting Gleason pattern 5 was 0·971 (0·943 to 0·998) in the external validation set. Cancer percentage calculated by pathologists and the algorithm showed good agreement (r=0·882, 95% CI 0·834 to 0·915; p<0·0001) with a mean bias of -4·14% (-6·36 to -1·91). The algorithm achieved an AUC of 0·957 (0·930 to 0·985) for perineural invasion. In routine practice, the algorithm was used to assess 11 429 H&E-stained slides pertaining to 941 cases leading to 90 Gleason score 7-10 alerts and 560 cancer alerts. 51 (9%) cancer alerts led to additional cuts or stains being ordered, two (4%) of which led to a third opinion request. We report on the first case of missed cancer that was detected by the algorithm.

INTERPRETATION

This study reports the successful development, external clinical validation, and deployment in clinical practice of an AI-based algorithm to accurately detect, grade, and evaluate clinically relevant findings in digitised slides of prostate CNBs.

FUNDING

Ibex Medical Analytics.