Number of pathologists in Germany: comparison with European countries, USA, and Canada

Joining forces for pathology diagnostics with AI assistance: The EMPAIA initiative

Over the past decade, artificial intelligence (AI) methods in pathology have advanced substantially. However, integration into routine clinical practice has been slow due to numerous challenges, including technical and regulatory hurdles in translating research results into clinical diagnostic products and the lack of standardized interfaces. The open and vendor-neutral EMPAIA initiative addresses these challenges. Here, we provide an overview of EMPAIA's achievements and lessons learned. EMPAIA integrates various stakeholders of the pathology AI ecosystem, i.e., pathologists, computer scientists, and industry. In close collaboration, we developed technical interoperability standards, recommendations for AI testing and product development, and explainability methods. We implemented the modular and open-source EMPAIA Platform and successfully integrated 14 AI-based image analysis apps from eight different vendors, demonstrating how different apps can use a single standardized interface. We prioritized requirements and evaluated the use of AI in real clinical settings with 14 different pathology laboratories in Europe and Asia. In addition to technical developments, we created a forum for all stakeholders to share information and experiences on digital pathology and AI. Commercial, clinical, and academic stakeholders can now adopt EMPAIA's common open-source interfaces, providing a unique opportunity for large-scale standardization and streamlining of processes. Further efforts are needed to effectively and broadly establish AI assistance in routine laboratory use. To this end, a sustainable infrastructure, the non-profit association EMPAIA International, has been established to continue standardization and support broad implementation and advocacy for an AI-assisted digital pathology future.

Detection of disease-specific signatures in B cell repertoires of lymphomas using machine learning

The classification of B cell lymphomas-mainly based on light microscopy evaluation by a pathologist-requires many years of training. Since the B cell receptor (BCR) of the lymphoma clonotype and the microenvironmental immune architecture are important features discriminating different lymphoma subsets, we asked whether BCR repertoire next-generation sequencing (NGS) of lymphoma-infiltrated tissues in conjunction with machine learning algorithms could have diagnostic utility in the subclassification of these cancers. We trained a random forest and a linear classifier via logistic regression based on patterns of clonal distribution, VDJ gene usage and physico-chemical properties of the top-n most frequently represented clonotypes in the BCR repertoires of 620 paradigmatic lymphoma samples-nodular lymphocyte predominant B cell lymphoma (NLPBL), diffuse large B cell lymphoma (DLBCL) and chronic lymphocytic leukemia (CLL)-alongside with 291 control samples. With regard to DLBCL and CLL, the models demonstrated optimal performance when utilizing only the most prevalent clonotype for classification, while in NLPBL-that has a dominant background of non-malignant bystander cells-a broader array of clonotypes enhanced model accuracy. Surprisingly, the straightforward logistic regression model performed best in this seemingly complex classification problem, suggesting linear separability in our chosen dimensions. It achieved a weighted F1-score of 0.84 on a test cohort including 125 samples from all three lymphoma entities and 58 samples from healthy individuals. Together, we provide proof-of-concept that at least the 3 studied lymphoma entities can be differentiated from each other using BCR repertoire NGS on lymphoma-infiltrated tissues by a trained machine learning model.

TelePi: an affordable telepathology microscope camera system anyone can build and use

Telepathology facilitates histological diagnoses through sharing expertise between pathologists. However, the associated costs are high and frequently prohibitive, especially in low-resource settings, where telepathology would paradoxically be of paramount importance due to a paucity of pathologists.We have constructed a telepathology system (TelePi) with a budget of < €120 using the small, single-board computer Raspberry Pi Zero and its High-Quality Camera Module in conjunction with a standard microscope and open-source software. The system requires no maintenance costs or service contracts, has a small footprint, can be moved and shared across several microscopes, and is independent from other computer operating systems. TelePi uses a responsive and high-resolution web-based live stream which allows remote consultation between two or more locations. TelePi can serve as a telepathology system for remote diagnostics of frozen sections. Additionally, it can be used as a standard microscope camera for teaching of medical students and for basic research. The quality of the TelePi system compared favorable to a commercially available telepathology system that exceed its cost by more than 125-fold. Additionally, still images are of publication quality equal to that of a whole slide scanner that costs 800 times more.In summary, TelePi is an affordable, versatile, and inexpensive camera system that potentially enables telepathology in low-resource settings without sacrificing image quality.

[Artificial intelligence in kidney transplant pathology]

BACKGROUND

Artificial intelligence (AI) systems have showed promising results in digital pathology, including digital nephropathology and specifically also kidney transplant pathology.

AIM

Summarize the current state of research and limitations in the field of AI in kidney transplant pathology diagnostics and provide a future outlook.

MATERIALS AND METHODS

Literature search in PubMed and Web of Science using the search terms "deep learning", "transplant", and "kidney". Based on these results and studies cited in the identified literature, a selection was made of studies that have a histopathological focus and use AI to improve kidney transplant diagnostics.

RESULTS AND CONCLUSION

Many studies have already made important contributions, particularly to the automation of the quantification of some histopathological lesions in nephropathology. This likely can be extended to automatically quantify all relevant lesions for a kidney transplant, such as Banff lesions. Important limitations and challenges exist in the collection of representative data sets and the updates of Banff classification, making large-scale studies challenging. The already positive study results make future AI support in kidney transplant pathology appear likely.

Potential rapid intraoperative cancer diagnosis using dynamic full-field optical coherence tomography and deep learning: A prospective cohort study in breast cancer patients

An intraoperative diagnosis is critical for precise cancer surgery. However, traditional intraoperative assessments based on hematoxylin and eosin (H&E) histology, such as frozen section, are time-, resource-, and labor-intensive, and involve specimen-consuming concerns. Here, we report a near-real-time automated cancer diagnosis workflow for breast cancer that combines dynamic full-field optical coherence tomography (D-FFOCT), a label-free optical imaging method, and deep learning for bedside tumor diagnosis during surgery. To classify the benign and malignant breast tissues, we conducted a prospective cohort trial. In the modeling group (n = 182), D-FFOCT images were captured from April 26 to June 20, 2018, encompassing 48 benign lesions, 114 invasive ductal carcinoma (IDC), 10 invasive lobular carcinoma, 4 ductal carcinoma in situ (DCIS), and 6 rare tumors. Deep learning model was built up and fine-tuned in 10,357 D-FFOCT patches. Subsequently, from June 22 to August 17, 2018, independent tests (n = 42) were conducted on 10 benign lesions, 29 IDC, 1 DCIS, and 2 rare tumors. The model yielded excellent performance, with an accuracy of 97.62%, sensitivity of 96.88% and specificity of 100%; only one IDC was misclassified. Meanwhile, the acquisition of the D-FFOCT images was non-destructive and did not require any tissue preparation or staining procedures. In the simulated intraoperative margin evaluation procedure, the time required for our novel workflow (approximately 3 min) was significantly shorter than that required for traditional procedures (approximately 30 min). These findings indicate that the combination of D-FFOCT and deep learning algorithms can streamline intraoperative cancer diagnosis independently of traditional pathology laboratory procedures.

Effectiveness and Cost-effectiveness of Artificial Intelligence-assisted Pathology for Prostate Cancer Diagnosis in Sweden: A Microsimulation Study

BACKGROUND AND OBJECTIVE

Image-based artificial intelligence (AI) methods have shown high accuracy in prostate cancer (PCa) detection. Their impact on patient outcomes and cost effectiveness in comparison to human pathologists remains unknown. Our aim was to evaluate the effectiveness and cost-effectiveness of AI-assisted pathology for PCa diagnosis in Sweden.

METHODS

We modeled quadrennial prostate-specific antigen (PSA) screening for men between the ages of 50 and 74 yr over a lifetime horizon using a health care perspective. Men with PSA ≥3 ng/ml were referred for standard biopsy (SBx), for which cores were either examined via AI followed by a pathologist for AI-labeled positive cores, or a pathologist alone. The AI performance characteristics were estimated using an internal STHLM3 validation data set. Outcome measures included the number of tests, PCa incidence and mortality, overdiagnosis, quality-adjusted life years (QALYs), and the potential reduction in pathologist-evaluated biopsy cores if AI were used. Cost-effectiveness was assessed using the incremental cost-effectiveness ratio.

KEY FINDINGS AND LIMITATIONS

In comparison to a pathologist alone, the AI-assisted workflow increased the number of PSA tests, SBx procedures, and PCa deaths by ≤0.03%, and slightly reduced PCa incidence and overdiagnosis. AI would reduce the proportion of biopsy cores evaluated by a pathologist by 80%. At a cost of €10 per case, the AI-assisted workflow would cost less and result in <0.001% lower QALYs in comparison to a pathologist alone. The results were sensitive to the AI cost.

CONCLUSIONS AND CLINICAL IMPLICATIONS

According to our model, AI-assisted pathology would significantly decrease the workload of pathologists, would not affect patient quality of life, and would yield cost savings in Sweden when compared to a human pathologist alone.

PATIENT SUMMARY

We compared outcomes for prostate cancer patients and relevant costs for two methods of assessing prostate biopsies in Sweden: (1) artificial intelligence (AI) technology and review of positive biopsies by a human pathologist; and (2) a human pathologist alone for all biopsies. We found that addition of AI would reduce the pathology workload and save money, and would not affect patient outcomes when compared to a human pathologist alone. The results suggest that adding AI to prostate pathology in Sweden would save costs.

Factors associated with interobserver variation amongst pathologists in the diagnosis of endometrial hyperplasia: A systematic review

OBJECTIVE

Reproducible diagnoses of endometrial hyperplasia (EH) remains challenging and has potential implications for patient management. This systematic review aimed to identify pathologist-specific factors associated with interobserver variation in the diagnosis and reporting of EH.

METHODS

Three electronic databases, namely MEDLINE, Embase and Web of Science, were searched from 1st January 2000 to 25th March 2023, using relevant key words and subject headings. Eligible studies reported on pathologist-specific factors or working practices influencing interobserver variation in the diagnosis of EH, using either the World Health Organisation (WHO) 2014 or 2020 classification or the endometrioid intraepithelial neoplasia (EIN) classification system. Quality assessment was undertaken using the QUADAS-2 tool, and findings were narratively synthesised.

RESULTS

Eight studies were identified. Interobserver variation was shown to be significant even amongst specialist gynaecological pathologists in most studies. Few studies investigated pathologist-specific characteristics, but pathologists were shown to have different diagnostic styles, with some more likely to under-diagnose and others likely to over-diagnose EH. Some novel working practices were identified, such as grading the "degree" of nuclear atypia and the incorporation of objective methods of diagnosis such as semi-automated quantitative image analysis/deep learning models.

CONCLUSIONS

This review highlighted the impact of pathologist-specific factors and working practices in the accurate diagnosis of EH, although few studies have been conducted. Further research is warranted in the development of more objective criteria that could improve reproducibility in EH diagnostic reporting, as well as determining the applicability of novel methods such as grading the degree of nuclear atypia in clinical settings.

[Digital transformation of a routine histopathology lab : Dos and don'ts!]

The implementation of digital histopathology in the laboratory marks a crucial milestone in the overall digital transformation of pathology. This shift offers a range of new possibilities, including access to extensive datasets for AI-assisted analyses, the flexibility of remote work and home office arrangements for specialists, and the expedited and simplified sharing of images and data for research, conferences, and tumor boards. However, the transition to a fully digital workflow involves significant technological and personnel-related efforts. It necessitates careful and adaptable change management to minimize disruptions, particularly in the personnel domain, and to prevent the loss of valuable potential from employees who may be resistant to change. This article consolidates our institute's experiences, highlighting technical and personnel-related challenges encountered during the transition to digital pathology. It also presents a comprehensive overview of potential difficulties at various interfaces when converting routine operations to a digital workflow.

[Effects of upstream laboratory processes on the digitization of histological slides]

BACKGROUND

Several factors in glass slide (GS) preparation affect the quality and data volume of a digitized histological slide. In particular, reducing contamination and selecting the appropriate coverslip have the potential to significantly reduce scan time and data volume.

GOALS

To objectify observations from our institute's digitization process to determine the impact of laboratory processes on the quality of digital histology slides.

MATERIALS AND METHODS

Experiment 1: Scanning the GS before and after installation of a central console in the microtomy area to reduce dirt and statistical analysis of the determined parameters. Experiment 2: Re-coverslipping the GS (post diagnostics) with glass and film. Scanning the GS and statistical analysis of the collected parameters.

CONCLUSION

The targeted restructuring in the laboratory process leads to a reduction of GS contamination. This causes a significant reduction in the amount of data generated and scanning time required for the digitized sections. Film as a coverslip material minimizes processing errors in contrast to glass. According to our estimation, all the above-mentioned points lead to considerable cost savings.

Comprehensive and Accurate Molecular Profiling of Breast Cancer through mRNA Expression of ESR1, PGR, ERBB2, MKI67, and a Novel Proliferation Signature

BACKGROUND

An accurate status determination of breast cancer biomarkers (ER, PR, HER2, Ki67) is crucial for guiding patient management. The "gold standard" for assessing these biomarkers in FFPE tissue is IHC, which faces challenges in standardization and exhibits substantial variability. In this study, we compare the concordance of a new commercial RT-qPCR kit with IHC in determining BC biomarker status.

METHODS

The performance was evaluated using 634 FFPE specimens, which underwent histological analysis in accordance with standard of care methods. HER2 2+ tumors were referred to ISH testing. An immunoreactive score of ≥2/12 was considered positive for ER/PR and 20% staining was used as a cut-off for Ki67 high/low score. RT-qPCR and results calling were performed according to the manufacturer's instructions.

RESULTS

High concordance with IHC was seen for all markers (93.2% for ER, 87.1% for PR, 93.9% for HER2, 77.9% for Ki67 and 80.1% for proliferative signature (assessed against Ki67 IHC)).

CONCLUSIONS

By assessing the concordance with the results obtained through IHC, we sought to demonstrate the reliability and utility of the kit for precise BC subtyping. Our findings suggest that the kit provides a highly precise and accurate quantitative assessment of BC biomarkers.

Operational greenhouse-gas emissions of deep learning in digital pathology: a modelling study

BACKGROUND

Deep learning is a promising way to improve health care. Image-processing medical disciplines, such as pathology, are expected to be transformed by deep learning. The first clinically applicable deep-learning diagnostic support tools are already available in cancer pathology, and their number is increasing. However, data on the environmental sustainability of these tools are scarce. We aimed to conduct an environmental-sustainability analysis of a theoretical implementation of deep learning in patient-care pathology.

METHODS

For this modelling study, we first assembled and calculated relevant data and parameters of a digital-pathology workflow. Data were breast and prostate specimens from the university clinic at the Institute of Pathology of the Rheinisch-Westfälische Technische Hochschule Aachen (Aachen, Germany), for which commercially available deep learning was already available. Only specimens collected between Jan 1 and Dec 31, 2019 were used, to omit potential biases due to the COVID-19 pandemic. Our final selection was based on 2 representative weeks outside holidays, covering different types of specimens. To calculate carbon dioxide (CO2) or CO2 equivalent (CO2 eq) emissions of deep learning in pathology, we gathered relevant data for exact numbers and sizes of whole-slide images (WSIs), which were generated by scanning histopathology samples of prostate and breast specimens. We also evaluated different data input scenarios (including all slide tiles, only tiles containing tissue, or only tiles containing regions of interest). To convert estimated energy consumption from kWh to CO2 eq, we used the internet protocol address of the computational server and the Electricity Maps database to obtain information on the sources of the local electricity grid (ie, renewable vs non-renewable), and estimated the number of trees and proportion of the local and world's forests needed to sequester the CO2 eq emissions. We calculated the computational requirements and CO2 eq emissions of 30 deep-learning models that varied in task and size. The first scenario represented the use of one commercially available deep-learning model for one task in one case (1-task), the second scenario considered two deep-learning models for two tasks per case (2-task), the third scenario represented a future, potentially automated workflow that could handle 7 tasks per case (7-task), and the fourth scenario represented the use of a single potential, large, computer-vision model that could conduct multiple tasks (multitask). We also compared the performance (ie, accuracy) and CO2 eq emissions of different deep-learning models for the classification of renal cell carcinoma on WSIs, also from Rheinisch-Westfälische Technische Hochschule Aachen. We also tested other approaches to reducing CO2 eq emissions, including model pruning and an alternative method for histopathology analysis (pathomics).

FINDINGS

The pathology database contained 35 552 specimens (237 179 slides), 6420 of which were prostate specimens (10 115 slides) and 11 801 of which were breast specimens (19 763 slides). We selected and subsequently digitised 140 slides from eight breast-cancer cases and 223 slides from five prostate-cancer cases. Applying large deep-learning models on all WSI tiles of prostate and breast pathology cases would result in yearly CO2 eq emissions of 7·65 metric tons (t; 95% CI 7·62-7·68) with the use of a single deep-learning model per case; yearly CO2 eq emissions were up to 100·56 t (100·21-100·99) with the use of seven deep-learning models per case. CO2 eq emissions for different deep-learning model scenarios, data inputs, and deep-learning model sizes for all slides varied from 3·61 t (3·59-3·63) to 2795·30 t (1177·51-6482·13. For the estimated number of overall pathology cases worldwide, the yearly CO2 eq emissions varied, reaching up to 16 megatons (Mt) of CO2 eq, requiring up to 86 590 km2 (0·22%) of world forest to sequester the CO2 eq emissions. Use of the 7-task scenario and small deep-learning models on slides containing tissue only could substantially reduce CO2 eq emissions worldwide by up to 141 times (0·1 Mt, 95% CI 0·1-0·1). Considering the local environment in Aachen, Germany, the maximum CO2 eq emission from the use of deep learning in digital pathology only would require 32·8% (95% CI 13·8-76·6) of the local forest to sequester the CO2 eq emissions. A single pathomics run on a tissue could provide information that was comparable to or even better than the output of multitask deep-learning models, but with 147 times reduced CO2 eq emissions.

INTERPRETATION

Our findings suggest that widespread use of deep learning in pathology might have considerable global-warming potential. The medical community, policy decision makers, and the public should be aware of this potential and encourage the use of CO2 eq emissions reduction strategies where possible.

FUNDING

German Research Foundation, European Research Council, German Federal Ministry of Education and Research, Health, Economic Affairs and Climate Action, and the Innovation Fund of the Federal Joint Committee.

Biomedical laboratory scientists and technicians in digital pathology - Is there a need for professional development?

Objective

Digital pathology (DP) is moving into Danish pathology departments at high pace. Conventionally, biomedical laboratory scientists (BLS) and technicians have prepared tissue sections for light microscopy, but workflow alterations are required for the new digital era with whole slide imaging (WSI); digitally assisted image analysis (DAIA) and artificial intelligence (AI). We aim to explore the role of BLS in DP and assess a potential need for professional development.

Methods

We investigated the roles of BLS in the new digital era through qualitative interviews at Danish Pathology Departments in 2019/2020 before DP implementation (supported by a questionnaire); and in 2022 after DP implementation. Additionally, senior lecturers from three Danish University Colleges reported on how DP was integrated into the 2023 bachelor's degree educational curricula for BLS students.

Results

At some Danish pathology departments, BLS were involved in the implementation process of DP and their greatest concerns were lack of physical laboratory requirements (69%) and implementation strategies (63%). BLS were generally positive towards working with DP, however, some expressed concern about extended working hours for scanning. Work-task transfers from pathologists were generally greeted positively from both management and pathologists; however, at follow-up interviews after DP implementation, job transfers had not been effectuated. At Danish university colleges, DP had been integrated systematically in the curricula for BLS students, especially WSI.

Conclusion

Involving BLS in DP implementation and development may benefit the process, as BLS have a hands-on workflow perspective with a focus on quality assurance. Several new work opportunities for BLS may occur with DP including WSI, DAIA and AI, and therefore new qualifications are warranted, which must be considered in future undergraduate programmes for BLS students or postgraduate programmes for BLS.

Looking beyond workforce parity: addressing gender inequity in pathology

While women pathologists have made up over one-third of pathologists in the Australian workforce for over 15 years and at least 50% since 2019, they are under-represented in senior leadership roles, scientific publications, grant recipients, editorial boards, key presentations, and professional awards. This is not unique to pathology and is seen in the broader medical and academic community. Barriers to gender equity and equality in pathology, medicine and academia include gender stereotypes, gender-based discrimination, structural and organisational barriers as well as broader social and cultural barriers. A diverse leadership reflective of the whole professional body and the broader community is important for optimal health outcomes. It is the responsibility and moral duty of individuals and organisations to address any gender disparities, inequities, and inequalities by monitoring, identifying, and acting on gender biases and systemic barriers that hinder appropriate levels of representation by women.

Implementation of deep learning in liver pathology optimizes diagnosis of benign lesions and adenocarcinoma metastasis

INTRODUCTION

Differentiation of histologically similar structures in the liver, including anatomical structures, benign bile duct lesions, or common types of liver metastases, can be challenging with conventional histological tissue sections alone. Accurate histopathological classification is paramount for the diagnosis and adequate treatment of the disease. Deep learning algorithms have been proposed for objective and consistent assessment of digital histopathological images.

MATERIALS AND METHODS

In the present study, we trained and evaluated deep learning algorithms based on the EfficientNetV2 and ResNetRS architectures to discriminate between different histopathological classes. For the required dataset, specialized surgical pathologists annotated seven different histological classes, including different non-neoplastic anatomical structures, benign bile duct lesions, and liver metastases from colorectal and pancreatic adenocarcinoma in a large patient cohort. Annotation resulted in a total of 204.159 image patches, followed by discrimination analysis using our deep learning models. Model performance was evaluated on validation and test data using confusion matrices.

RESULTS

Evaluation of the test set based on tiles and cases revealed overall highly satisfactory prediction capability of our algorithm for the different histological classes, resulting in a tile accuracy of 89% (38 413/43 059) and case accuracy of 94% (198/211). Importantly, the separation of metastasis versus benign lesions was certainly confident on case level, confirming the classification model performed with high diagnostic accuracy. Moreover, the whole curated raw data set is made publically available.

CONCLUSIONS

Deep learning is a promising approach in surgical liver pathology supporting decision making in personalized medicine.

Perception of Pathologists in Poland of Artificial Intelligence and Machine Learning in Medical Diagnosis-A Cross-Sectional Study

BACKGROUND

In the past vicennium, several artificial intelligence (AI) and machine learning (ML) models have been developed to assist in medical diagnosis, decision making, and design of treatment protocols. The number of active pathologists in Poland is low, prolonging tumor patients' diagnosis and treatment journey. Hence, applying AI and ML may aid in this process. Therefore, our study aims to investigate the knowledge of using AI and ML methods in the clinical field in pathologists in Poland. To our knowledge, no similar study has been conducted.

METHODS

We conducted a cross-sectional study targeting pathologists in Poland from June to July 2022. The questionnaire included self-reported information on AI or ML knowledge, experience, specialization, personal thoughts, and level of agreement with different aspects of AI and ML in medical diagnosis. Data were analyzed using IBM^® SPSS^® Statistics v.26, PQStat Software v.1.8.2.238, and RStudio Build 351.

RESULTS

Overall, 68 pathologists in Poland participated in our study. Their average age and years of experience were 38.92 ± 8.88 and 12.78 ± 9.48 years, respectively. Approximately 42% used AI or ML methods, which showed a significant difference in the knowledge gap between those who never used it (OR = 17.9, 95% CI = 3.57-89.79, p < 0.001). Additionally, users of AI had higher odds of reporting satisfaction with the speed of AI in the medical diagnosis process (OR = 4.66, 95% CI = 1.05-20.78, p = 0.043). Finally, significant differences (p = 0.003) were observed in determining the liability for legal issues used by AI and ML methods.

CONCLUSION

Most pathologists in this study did not use AI or ML models, highlighting the importance of increasing awareness and educational programs regarding applying AI and ML in medical diagnosis.

Shortage of pathologists in Italy: survey of students and residents

Objective

In Italy, shortage of pathologists is a problem that affects the quality of the National Health System (NHS). The cause of the shortage of pathologists in Italy must be sought in the lack of interests in the pathologist career by Medical Course Students (MCS) and in drop out of Post-Graduate Medical Schools (PGMS). We investigated reasons of both through two surveys.

Methods

We developed and proposed on Facebook two surveys, one to MCSs attending last years of study and one to Pathology School Residents (PSRs). Survey for MCSs consisted of 10 questions centered on their perception about pathologist activity; survey for PSRs consisted of 8 questions and investigated the most and least appreciated aspects of Italian PGMS.

Results

We obtained 500 responses from the MCSs and 51 responses from the PSRs. Our results show that lack of interest of MCS may be due to their incomplete knowledge of the pathologist's activities. On the other hand, PSR answers show that some teaching aspects should be improved.

Conclusions

Our surveys showed that lack of interest of MCS in the pathology career depends on poor knowledge about the real clinical significance of pathology and PSRs believe that Italian PGMS do not meet their interest. One solution could be a renewal of teaching both in the pathology courses for MCS and in PGMS.

Using less annotation workload to establish a pathological auxiliary diagnosis system for gastric cancer

Pathological diagnosis of gastric cancer requires pathologists to have extensive clinical experience. To help pathologists improve diagnostic accuracy and efficiency, we collected 1,514 cases of stomach H&E-stained specimens with complete diagnostic information to establish a pathological auxiliary diagnosis system based on deep learning. At the slide level, our system achieves a specificity of 0.8878 while maintaining a high sensitivity close to 1.0 on 269 biopsy specimens (147 malignancies) and 163 surgical specimens (80 malignancies). The classified accuracy of our system is 0.9034 at the slide level for 352 biopsy specimens (201 malignancies) from 50 medical centers. With the help of our system, the pathologists' average false-negative rate and average false-positive rate on 100 biopsy specimens (50 malignancies) are reduced to 1/5 and 1/2 of the original rates, respectively. At the same time, the average uncertainty rate and the average diagnosis time are reduced by approximately 22% and 20%, respectively.

Mapping the current state of the medical specialties in laboratory medicine in Italy

Laboratory medicine is the single highest volume medical activity and it plays an increasingly essential role in the modern healthcare systems. In vitro diagnostic tests are now used in virtually every aspect of the patient care, including disease prevention, diagnosis, disease monitoring as well as personalised medicine. Nevertheless, the profession of laboratory medicine differs between countries in many respects, such as specialist training (medical or scientific), fields of interest, responsibilities and professional organisation. Many attempts have been made to quantify the role of laboratory medicine in patient outcomes, but the precise figures are still not clear. Moreover, the relative contribution of medical specialists in laboratory medicine is not well known and somehow controversial. To start exploring these aspects, we studied the current state of the two medical specialties that make up the majority of laboratory medicine in Italy: clinical pathology and medical microbiology. Our analysis revealed that both specialties suffer from a low attractivity among postgraduate physicians, and suggest that a restructuring of the training programme and professional reorganisation should be considered.

Artificial intelligence and digital biomarker in precision pathology guiding immune therapy selection and precision oncology

BACKGROUND

The currently available immunotherapies already changed the strategy how many cancers are treated from first to last line. Understanding even the most complex heterogeneity in tumor tissue and mapping the spatial cartography of the tumor immunity allows the best and optimized selection of immune modulating agents to (re-)activate the patient's immune system and direct it against the individual cancer in the most effective way.

RECENT FINDINGS

Primary cancer and metastases maintain a high degree of plasticity to escape any immune surveillance and continue to evolve depending on many intrinsic and extrinsic factors In the field of immune-oncology (IO) immune modulating agents are recognized as practice changing therapeutic modalities. Recent studies have shown that an optimal and lasting efficacy of IO therapeutics depends on the understanding of the spatial communication network and functional context of immune and cancer cells within the tumor microenvironment. Artificial intelligence (AI) provides an insight into the immune-cancer-network through the visualization of very complex tumor and immune interactions in cancer tissue specimens and allows the computer-assisted development and clinical validation of such digital biomarker.

CONCLUSIONS

The successful implementation of AI-supported digital biomarker solutions guides the clinical selection of effective immune therapeutics based on the retrieval and visualization of spatial and contextual information from cancer tissue images and standardized data. As such, computational pathology (CP) turns into "precision pathology" delivering individual therapy response prediction. Precision Pathology does not only include digital and computational solutions but also high levels of standardized processes in the routine histopathology workflow and the use of mathematical tools to support clinical and diagnostic decisions as the basic principle of a "precision oncology".

Characterising the use of surgical pathology rush requests: a descriptive analysis and survey

This study aimed to characterise priority or 'rush' surgical pathology requests and identify potentially targetable factors. We performed a retrospective descriptive analysis of rush requests at our institution from 2016 to 2019 and conducted a survey asking pathologists about their perspectives on rush cases. There were 3677 rush cases, with case characteristics generally stable over the study period. Two categories of requests were identified based on hospital status; outpatient requests more frequently provided a specific date for diagnosis, while inpatient rush requests generally required a diagnosis as soon as possible. Most pathologists found rush cases to be somewhat more stressful compared with routine cases (65.2%) and found it very or extremely useful to know when a result is needed (86.9%). The use of hospitalisation status, and identifying if results are required by a certain date, may help in more effective triaging of rush surgical pathology cases.

Quality Assessment of Cervical Cytology Practices in Estonia From 2007 to 2018

BACKGROUND

Cervical cancer incidence and mortality in Estonia are among the highest in Europe, although the overall coverage with cervical cytology is high. This indicates potential issues with the quality of collection and/or laboratory evaluation of cervical cytology.

OBJECTIVES

The aim of the retrospective observational study was to assess the quality of cervical cytology specimen collection, evaluation, and reporting using laboratory reports in Estonia.

METHODS

The study included women with a cervical cancer diagnosis in 2017-2018. Cervical cytology and histology reports for these women in 2007-2018 were obtained from ten laboratories. We described the quality of cytology specimen collection and reporting of cytology results. Multivariate logistic regression was used to calculate odds ratios (OR) with 95% confidence intervals (CI) to identify factors associated with NILM as the last cervical cytology result within 5 or 2 years before the cervical cancer diagnosis. Also, we calculated cytology-histology correlation (CHC).

RESULTS

We identified 503 cytology and 100 histology reports from 138 women. The laboratories differed greatly regarding human resources, work capacity and volume. Differences between local and regional laboratories were observed in reporting specimen adequacy (P < .001). We found that local laboratories had 3 times higher odds (OR = 2.95, 95% CI: 1.05-8.33) of reporting normal results 2 years before cancer diagnosis than regional laboratories. According to the CHC, 58.9% of pairs were in agreement.

CONCLUSIONS

The study showed considerable heterogeneity and suboptimal performance of cervical cytology practices in Estonia, particularly at local laboratories. Efforts to improve laboratory quality assurance are crucial.

Validation and real-world clinical application of an artificial intelligence algorithm for breast cancer detection in biopsies

Breast cancer is the most common malignant disease worldwide, with over 2.26 million new cases in 2020. Its diagnosis is determined by a histological review of breast biopsy specimens, which can be labor-intensive, subjective, and error-prone. Artificial Intelligence (AI)-based tools can support cancer detection and classification in breast biopsies ensuring rapid, accurate, and objective diagnosis. We present here the development, external clinical validation, and deployment in routine use of an AI-based quality control solution for breast biopsy review. The underlying AI algorithm is trained to identify 51 different types of clinical and morphological features, and it achieves very high accuracy in a large, multi-site validation study. Specifically, the area under the receiver operating characteristic curves (AUC) for the detection of invasive carcinoma and of ductal carcinoma in situ (DCIS) are 0.99 (specificity and sensitivity of 93.57 and 95.51%, respectively) and 0.98 (specificity and sensitivity of 93.79 and 93.20% respectively), respectively. The AI algorithm differentiates well between subtypes of invasive and different grades of in situ carcinomas with an AUC of 0.97 for invasive ductal carcinoma (IDC) vs. invasive lobular carcinoma (ILC) and AUC of 0.92 for DCIS high grade vs. low grade/atypical ductal hyperplasia, respectively, as well as accurately identifies stromal tumor-infiltrating lymphocytes (TILs) with an AUC of 0.965. Deployment of this AI solution as a real-time quality control solution in clinical routine leads to the identification of cancers initially missed by the reviewing pathologist, demonstrating both clinical utility and accuracy in real-world clinical application.

Deep learning for the detection of anatomical tissue structures and neoplasms of the skin on scanned histopathological tissue sections

Basal cell carcinoma (BCC), squamous cell carcinoma (SqCC) and melanoma are among the most common cancer types. Correct diagnosis based on histological evaluation after biopsy or excision is paramount for adequate therapy stratification. Deep learning on histological slides has been suggested to complement and improve routine diagnostics, but publicly available curated and annotated data and usable models trained to distinguish common skin tumors are rare and often lack heterogeneous non-tumor categories. A total of 16 classes from 386 cases were manually annotated on scanned histological slides, 129,364 100 x 100 µm (~395 x 395 px) image tiles were extracted and split into a training, validation and test set. An EfficientV2 neuronal network was trained and optimized to classify image categories. Cross entropy loss, balanced accuracy and Matthews correlation coefficient were used for model evaluation. Image and patient data were assessed with confusion matrices. Application of the model to an external set of whole slides facilitated localization of melanoma and non-tumor tissue. Automated differentiation of BCC, SqCC, melanoma, naevi and non-tumor tissue structures was possible, and a high diagnostic accuracy was achieved in the validation (98%) and test (97%) set. In summary, we provide a curated dataset including the most common neoplasms of the skin and various anatomical compartments to enable researchers to train, validate and improve deep learning models. Automated classification of skin tumors by deep learning techniques is possible with high accuracy, facilitates tumor localization and has the potential to support and improve routine diagnostics.

[Digitalization in dermatopathology]

The histomorphological analysis of tissue sections by specially trained dermatopathologists is a central component for making the dermatological diagnosis. It is the foundation for the understanding of clinical aspects, pathophysiology and not least the treatment of skin diseases and is therefore an essential part of modern dermatology. New technological developments in recent years offer a variety of possibilities to digitalize dermatopathology, which could significantly change and even revolutionize the work of dermatopathologists in the coming years; however, like any new development there are limiting factors and open questions that need to be discussed. This article is intended to provide an overview of the current state of the art and to highlight the corresponding opportunities and risks on the road to digital dermatopathology.

Systemic anti-cancer therapy patterns in advanced non-small cell lung cancer in Europe

BACKGROUND

Systemic anti-cancer therapy (SACT) is the recommended treatment modality in patients with advanced non-small cell lung cancer (aNSCLC) in clinical guidelines. SACT options in aNSCLC have multiplied in recent years with the introduction of immunotherapy and targeted therapy. This article presents findings from the first comparative analysis of SACT patterns in Europe.

METHODS

SACT rates in aNSCLC were estimated as the ratio between the number of patients treated with SACT (chemotherapy, immunotherapy, targeted therapy) and the number of potentially eligible patients for SACT in 11 countries (Belgium, Bulgaria, Finland, Hungary, Ireland, Netherlands, Norway, Poland, Portugal, Romania, UK) between 2014 and 2020. Treated patients were estimated by combining national sales volume data of cancer drugs and average drug use per patient based on clinical trials. Potentially eligible patients were estimated from national epidemiological data.

RESULTS

SACT rates in aNSCLC differed greatly, ranging from around 30 % in Hungary, Poland, and the UK to almost 60 % in Ireland, Norway, and Portugal in 2014. SACT rates seemed to increase over time in most countries, but differences were still large by 2020, ranging from around 40 % in the UK to 75 % or more in Belgium, Norway, and Portugal. Even in countries with the highest SACT rates, far from all patients seemed to receive guideline-recommended SACT options, as underuse of immunotherapy and targeted therapy was common.

CONCLUSION

Up to 35 % of eligible patients with aNSCLC receives no SACT in certain European countries, although improvements have been achieved over time. The use of immunotherapy and targeted therapy is suboptimal even in countries with high SACT rates, indicating room to improve the quality of care and patient outcomes.

POLICY SUMMARY

Measuring if and what kind of therapy cancer patients have access to is vital to assess quality of care. The care of aNSCLC patients seems to be suboptimal in Europe, due to factors such as exclusion of patients with moderate performance status from SACT, limited resources for diagnostic testing, long reimbursement timelines and slow adoption of new medicines in clinical practice.

Automatische Bildanalyse und künstliche Intelligenz in der Nephropathologie

Hintergrund

Fragestellung

Material und Methoden

Ergebnisse

Diskussion

[Automation and the use of robots in the pathology laboratory : A journey through time and a consideration of efficiency]

In the past 20 years, numerous technical innovations were introduced to the histopathology laboratory, providing tools for improved standardization and occupational safety. Digital tracking serves as a backbone accompanying the workflow from the labeling of cassettes and slides to the final steps of preparing whole slide images and archiving blocks and sections. Multifunctional devices eliminated time consuming manual work prone to mistakes and loss of materials. At present, collaborative robots take over manual work that was considered to be exclusive to humans. The advent of these new technologies is expected to ameliorate the increasing staffing shortage in the laboratory but also of histopathologists.

[A European comparison of continuing education in pathology]

In the coming years, the shortage of pathologists will become internationally evident. In addition, the increase in knowledge, technical transformation processes, and the attractiveness of working conditions pose clear challenges for the field of pathology. A bi-directional opening for international mobility of pathologists could be a potential solution.In this analysis, the European training concept of the European Union of Medical Specialists (UEMS) was compared with its implementation in the 27 countries of the EU plus its 4 associated countries with regard to nationally differentiated concepts, type and implementation of the specialist examination, and additional qualifications. Subsequently, questions regarding the recognition of exams, titles, and specialist exams were elicited.The duration of training ranges between 4 and 6 years. The number of cases also varies considerably. Obtaining the specialist title can be done by simply completing the specifications up to a structured examination. In the EU, exams are mutually recognized, but this does not necessarily apply to academic titles and additional qualifications. Increasingly, on-site training centers are also subject to auditing procedures.The European agreements allow a high degree of permeability. However, national regulations pose hurdles for international mobility. The UEMS is therefore focusing on harmonization, including the certification of training centers. The so-called European Pathology Progress Test of the European Society of Pathology (ESP) is a further step towards the development of a future European specialist title. It remains the joint responsibility of residents and institutes to shape the future of the next generation of pathologists from the variety of different concepts.

Mass Spectrometry Imaging Spatial Tissue Analysis toward Personalized Medicine

Novel profiling methodologies are redefining the diagnostic capabilities and therapeutic approaches towards more precise and personalized healthcare. Complementary information can be obtained from different omic approaches in combination with the traditional macro- and microscopic analysis of the tissue, providing a more complete assessment of the disease. Mass spectrometry imaging, as a tissue typing approach, provides information on the molecular level directly measured from the tissue. Lipids, metabolites, glycans, and proteins can be used for better understanding imbalances in the DNA to RNA to protein translation, which leads to aberrant cellular behavior. Several studies have explored the capabilities of this technology to be applied to tumor subtyping, patient prognosis, and tissue profiling for intraoperative tissue evaluation. In the future, intercenter studies may provide the needed confirmation on the reproducibility, robustness, and applicability of the developed classification models for tissue characterization to assist in disease management.

Artificial Intelligence in Pediatric Pathology: The Extinction of a Medical Profession or the Key to a Bright Future?

Artificial Intelligence (AI) has become of increasing interest over the past decade. While digital image analysis (DIA) is already being used in radiology, it is still in its infancy in pathology. One of the reasons is that large-scale digitization of glass slides has only recently become available. With the advent of digital slide scanners, that digitize glass slides into whole slide images, many labs are now in a transition phase towards digital pathology. However, only few departments worldwide are currently fully digital. Digital pathology provides the ability to annotate large datasets and train computers to develop and validate robust algorithms, similar to radiology. In this opinionated overview, we will give a brief introduction into AI in pathology, discuss the potential positive and negative implications and speculate about the future role of AI in the field of pediatric pathology.

Validation of whole slide imaging for frozen section diagnosis of lymph node metastasis: A retrospective study from a tertiary care hospital in Thailand

BACKGROUND

The use of whole slide imaging (WSI) for frozen section (FS) diagnosis is helpful, particularly in the context of pathologist shortages. However, there is minimal data on such usage in resource-limited settings. This study aims to validate the use of WSI for FS diagnosis of lymph node metastasis using a low-cost virtual microscope scanner with consumer-grade laptops at a tertiary care hospital in Thailand.

METHODS

FS slides were retrieved for which the clinical query was to evaluate lymph node metastasis. They were digitized by a virtual microscope scanner (MoticEasyScan, Hong Kong) using up to 40× optical magnification. Three observers with different pathology experience levels diagnosed each slide, reviewing glass slides (GS) followed by digital slides (DS) after two weeks of a wash out period. WSI and GS diagnoses were compared. The time used for scanning and diagnosis of each slide was recorded.

RESULTS

295 FS slides were retrieved and digitized. The first-time successful scanning rate was 93.6 %. The mean scanning time was 2 min per slide. Both intraobserver agreement and interobserver agreement of WSI and GS diagnoses were high (Cohen's K; kappa value >0.84). The time used for DS diagnosis decreased as the observer's experience with WSI increased.

CONCLUSIONS

Despite varying pathological experiences, observers using WSI provided accurate FS diagnoses of lymph node metastasis. The time required for DS diagnoses decreased with additional observer's experience with WSI. Therefore, a WSI system containing low-cost scanners and consumer-grade laptops could be used for FS services in hospital laboratories lacking pathologists.

Development of an intraoperative breast cancer margin assessment method using quantitative fluorescence measurements

Breast-conserving surgery has become the preferred treatment method for breast cancer. Surgical margin assessment is performed during surgery, as it can reduce local recurrence in the preserved breast. Development of reliable and lower-cost ex vivo cancer detection methods would offer several benefits for patient care. Here, a practical and quantitative evaluation method for the ex vivo fluorescent diagnosis of breast lesions was developed and confirmed through a three-step clinical study. Gamma-glutamyl-hydroxymethyl rhodamine green (gGlu-HMRG) has been reported to generate fluorescence in breast lesions. Using this probe, we constructed a reliable and reproducible procedure for the quantitative evaluation of fluorescence levels. We evaluated the reliability of the method by considering reproducibility, temperature sensitivity, and the effects of other clinicopathological factors. The results suggest that the fluorescence increase of gGlu-HMRG is a good indicator of the malignancy of breast lesions. However, the distributions overlapped. A 5 min reaction with this probe could be used to distinguish at least part of the normal breast tissue. This method did not affect the final pathological examination. In summary, our results indicate that the methods developed in this study may serve as a feasible intraoperative negative-margin assessment tool during breast-conserving surgery.

[Automation and application of robotics in the pathology laboratory]

Over the last 20 years, numerous technical innovations have been introduced to the histopathology laboratory, providing tools for improved standardization and occupational safety. Digital tracking serves as a backbone accompanying the workflow from labeling cassettes and slides to the final steps of preparation of whole slide images and archiving blocks and sections. Multifunctional devices eliminated time consuming manual work prone to mistakes and loss of materials. At present, collaborative robots take over manual work that was considered to be exclusive to humans. The advent of these new technologies is expected to ameliorate the increasing staffing shortage in the laboratory and on the side of histopathologists as well.

Renal Cancer Detection: Fusing Deep and Texture Features from Histopathology Images

Histopathological images contain morphological markers of disease progression that have diagnostic and predictive values, with many computer-aided diagnosis systems using common deep learning methods that have been proposed to save time and labour. Even though deep learning methods are an end-to-end method, they perform exceptionally well given a large dataset and often show relatively inferior results for a small dataset. In contrast, traditional feature extraction methods have greater robustness and perform well with a small/medium dataset. Moreover, a texture representation-based global approach is commonly used to classify histological tissue images expect in explicit segmentation to extract the structure properties. Considering the scarcity of medical datasets and the usefulness of texture representation, we would like to integrate both the advantages of deep learning and traditional machine learning, i.e., texture representation. To accomplish this task, we proposed a classification model to detect renal cancer using a histopathology dataset by fusing the features from a deep learning model with the extracted texture feature descriptors. Here, five texture feature descriptors from three texture feature families were applied to complement Alex-Net for the extensive validation of the fusion between the deep features and texture features. The texture features are from (1) statistic feature family: histogram of gradient, gray-level cooccurrence matrix, and local binary pattern; (2) transform-based texture feature family: Gabor filters; and (3) model-based texture feature family: Markov random field. The final experimental results for classification outperformed both Alex-Net and a singular texture descriptor, showing the effectiveness of combining the deep features and texture features in renal cancer detection.

MIXTURE of human expertise and deep learning-developing an explainable model for predicting pathological diagnosis and survival in patients with interstitial lung disease

Interstitial pneumonia is a heterogeneous disease with a progressive course and poor prognosis, at times even worse than those in the main cancer types. Histopathological examination is crucial for its diagnosis and estimation of prognosis. However, the evaluation strongly depends on the experience of pathologists, and the reproducibility of diagnosis is low. Herein, we propose MIXTURE (huMan-In-the-loop eXplainable artificial intelligence Through the Use of REcurrent training), an original method to develop deep learning models for extracting pathologically significant findings based on an expert pathologist's perspective with a small annotation effort. The procedure of MIXTURE consists of three steps as follows. First, we created feature extractors for tiles from whole slide images using self-supervised learning. The similar looking tiles were clustered based on the output features and then pathologists integrated the pathologically synonymous clusters. Using the integrated clusters as labeled data, deep learning models to classify the tiles into pathological findings were created by transfer-learning the feature extractors. We developed three models for different magnifications. Using these extracted findings, our model was able to predict the diagnosis of usual interstitial pneumonia, a finding suggestive of progressive disease, with high accuracy (AUC 0.90 in validation set and AUC 0.86 in test set). This high accuracy could not be achieved without the integration of findings by pathologists. The patients predicted as UIP had poorer prognosis (5-year overall survival [OS]: 55.4%) than those predicted as non-UIP (OS: 95.2%). The Cox proportional hazards model for each microscopic finding and prognosis pointed out dense fibrosis, fibroblastic foci, elastosis, and lymphocyte aggregation as independent risk factors. We suggest that MIXTURE may serve as a model approach to different diseases evaluated by medical imaging, including pathology and radiology, and be the prototype for explainable artificial intelligence that can collaborate with humans.

Automatic breast carcinoma detection in histopathological micrographs based on Single Shot Multibox Detector

Background

A diagnosis with histological classification by pathologists is very important for appropriate treatments to improve the prognosis of patients with breast cancer. However, the number of pathologists is limited, and assisting the pathological diagnosis by artificial intelligence becomes very important. Here, we presented an automatic breast lesions detection model using microscopic histopathological images based on a Single Shot Multibox Detector (SSD) for the first time and evaluated its significance in assisting the diagnosis.

Methods

We built the data set and trained the SSD model with 1361 microscopic images and evaluated using 315 images. Pathologists and medical students diagnosed the images with or without the assistance of the model to investigate the significance of our model in assisting the diagnosis.

Results

The model achieved 88.3% and 90.5% diagnostic accuracies in 3-class (benign, non-invasive carcinoma, or invasive carcinoma) or 2-class (benign or malignant) classification tasks, respectively, and the mean intersection over union was 0.59. Medical students achieved a remarkably higher diagnostic accuracy score (average 84.7%) with the assistance of the model compared to those without assistance (average 67.4%). Some people diagnosed images in a short time using the assistance of the model (shorten by average 6.4 min) while others required a longer time (extended by 7.2 min).

Conclusion

We presented the automatic breast lesions detection method at high speed using histopathological micrographs. The present system may conveniently support the histological diagnosis by pathologists in laboratories.

Automated assessment of Ki-67 proliferation index in neuroendocrine tumors by deep learning

The Ki‐67 proliferation index (PI) is a prognostic factor in neuroendocrine tumors (NETs) and defines tumor grade. Analysis of Ki‐67 PI requires calculation of Ki‐67‐positive and Ki‐67‐negative tumor cells, which is highly subjective. To overcome this, we developed a deep learning‐based Ki‐67 PI algorithm (KAI) that objectively calculates Ki‐67 PI. Our study material consisted of NETs divided into training (n = 39), testing (n = 124), and validation (n = 60) series. All slides were digitized and processed in the Aiforia^® Create (Aiforia Technologies, Helsinki, Finland) platform. The ICC between the pathologists and the KAI was 0.89. In 46% of the tumors, the Ki‐67 PIs calculated by the pathologists and the KAI were the same. In 12% of the tumors, the Ki‐67 PI calculated by the KAI was 1% lower and in 42% of the tumors on average 3% higher. The DL‐based Ki‐67 PI algorithm yields results similar to human observers. While the algorithm cannot replace the pathologist, it can assist in the laborious Ki‐67 PI assessment of NETs. In the future, this approach could be useful in, for example, multi‐center clinical trials where objective estimation of Ki‐67 PI is crucial.

Deep Learning of Histopathological Features for the Prediction of Tumour Molecular Genetics

Advanced diagnostics are enabling cancer treatments to become increasingly tailored to the individual through developments in immunotherapies and targeted therapies. However, long turnaround times and high costs of molecular testing hinder the widespread implementation of targeted cancer treatments. Meanwhile, gold-standard histopathological assessment carried out by a trained pathologist is widely regarded as routine and mandatory in most cancers. Recently, methods have been developed to mine hidden information from histopathological slides using deep learning applied to scanned and digitized slides; deep learning comprises a collection of computational methods which learn patterns in data in order to make predictions. Such methods have been reported to be successful in a variety of cancers for predicting the presence of biomarkers such as driver mutations, tumour mutational burden, and microsatellite instability. This information could prove valuable to pathologists and oncologists in clinical decision making for cancer treatment and triage for in-depth sequencing. In addition to identifying molecular features, deep learning has been applied to predict prognosis and treatment response in certain cancers. Despite reported successes, many challenges remain before the clinical implementation of such diagnostic strategies in the clinical setting is possible. This review aims to outline recent developments in the field of deep learning for predicting molecular genetics from histopathological slides, as well as to highlight limitations and pitfalls of working with histopathology slides in deep learning.

Deep Learning in Pancreatic Tissue: Identification of Anatomical Structures, Pancreatic Intraepithelial Neoplasia, and Ductal Adenocarcinoma

Identification of pancreatic ductal adenocarcinoma (PDAC) and precursor lesions in histological tissue slides can be challenging and elaborate, especially due to tumor heterogeneity. Thus, supportive tools for the identification of anatomical and pathological tissue structures are desired. Deep learning methods recently emerged, which classify histological structures into image categories with high accuracy. However, to date, only a limited number of classes and patients have been included in histopathological studies. In this study, scanned histopathological tissue slides from tissue microarrays of PDAC patients (n = 201, image patches n = 81.165) were extracted and assigned to a training, validation, and test set. With these patches, we implemented a convolutional neuronal network, established quality control measures and a method to interpret the model, and implemented a workflow for whole tissue slides. An optimized EfficientNet algorithm achieved high accuracies that allowed automatically localizing and quantifying tissue categories including pancreatic intraepithelial neoplasia and PDAC in whole tissue slides. SmoothGrad heatmaps allowed explaining image classification results. This is the first study that utilizes deep learning for automatic identification of different anatomical tissue structures and diseases on histopathological images of pancreatic tissue specimens. The proposed approach is a valuable tool to support routine diagnostic review and pancreatic cancer research.

Deep Learning for the Classification of Non-Hodgkin Lymphoma on Histopathological Images

Simple Summary

Histopathological examination of lymph node (LN) specimens allows the detection of hematological diseases. The identification and the classification of lymphoma, a blood cancer with a manifestation in LNs, are difficult and require many years of training, as well as additional expensive investigations. Today, artificial intelligence (AI) can be used to support the pathologist in identifying abnormalities in LN specimens. In this article, we trained and optimized an AI algorithm to automatically detect two common lymphoma subtypes that require different therapies using normal LN parenchyma as a control. The balanced accuracy in an independent test cohort was above 95%, which means that the vast majority of cases were classified correctly and only a few cases were misclassified. We applied specific methods to explain which parts of the image were important for the AI algorithm and to ensure a reliable result. Our study shows that classifications of lymphoma subtypes is possible with high accuracy. We think that routine histopathological applications for AI should be pursued.

Abstract

The diagnosis and the subtyping of non-Hodgkin lymphoma (NHL) are challenging and require expert knowledge, great experience, thorough morphological analysis, and often additional expensive immunohistological and molecular methods. As these requirements are not always available, supplemental methods supporting morphological-based decision making and potentially entity subtyping are required. Deep learning methods have been shown to classify histopathological images with high accuracy, but data on NHL subtyping are limited. After annotation of histopathological whole-slide images and image patch extraction, we trained and optimized an EfficientNet convolutional neuronal network algorithm on 84,139 image patches from 629 patients and evaluated its potential to classify tumor-free reference lymph nodes, nodal small lymphocytic lymphoma/chronic lymphocytic leukemia, and nodal diffuse large B-cell lymphoma. The optimized algorithm achieved an accuracy of 95.56% on an independent test set including 16,960 image patches from 125 patients after the application of quality controls. Automatic classification of NHL is possible with high accuracy using deep learning on histopathological images and routine diagnostic applications should be pursued.

Artificial intelligence and pathology: From principles to practice and future applications in histomorphology and molecular profiling

The complexity of diagnostic (surgical) pathology has increased substantially over the last decades with respect to histomorphological and molecular profiling. Pathology has steadily expanded its role in tumor diagnostics and beyond from disease entity identification via prognosis estimation to precision therapy prediction. It is therefore not surprising that pathology is among the disciplines in medicine with high expectations in the application of artificial intelligence (AI) or machine learning approaches given their capabilities to analyze complex data in a quantitative and standardized manner to further enhance scope and precision of diagnostics. While an obvious application is the analysis of histological images, recent applications for the analysis of molecular profiling data from different sources and clinical data support the notion that AI will enhance both histopathology and molecular pathology in the future. At the same time, current literature should not be misunderstood in a way that pathologists will likely be replaced by AI applications in the foreseeable future. Although AI will transform pathology in the coming years, recent studies reporting AI algorithms to diagnose cancer or predict certain molecular properties deal with relatively simple diagnostic problems that fall short of the diagnostic complexity pathologists face in clinical routine. Here, we review the pertinent literature of AI methods and their applications to pathology, and put the current achievements and what can be expected in the future in the context of the requirements for research and routine diagnostics.

Breast cancer pathology services in sub-Saharan Africa: a survey within population-based cancer registries

BACKGROUND

Pathologists face major challenges in breast cancer diagnostics in sub-Saharan Africa (SSA). The major problems identified as impairing the quality of pathology reports are shortcomings of equipment, organization and insufficiently qualified personnel. In addition, in the context of breast cancer, immunohistochemistry (IHC) needs to be available for the evaluation of biomarkers. In the study presented, we aim to describe the current state of breast cancer pathology in order to highlight the unmet needs.

METHODS

We obtained information on breast cancer pathology services within population-based cancer registries in SSA. A survey of 20 participating pathology centres was carried out. These centres represent large, rather well-equipped pathologies. The data obtained were related to the known population and breast cancer incidence of the registry areas.

RESULTS

The responding pathologists served populations of between 30,000 and 1.8 million and the centres surveyed dealt with 10-386 breast cancer cases per year. Time to fixation and formalin fixation time varied from overnight to more than 72 h. Only five centres processed core needle biopsies as a daily routine. Technical problems were common, with 14 centres reporting temporary power outages and 18 centres claiming to own faulty equipment with no access to technical support. Only half of the centres carried out IHC in their own laboratory. For three centres, IHC was only accessible outside of the country and one centre could not obtain any IHC results. A tumour board was established in 13 centres.

CONCLUSIONS

We conclude that breast cancer pathology services ensuring state-of-the-art therapy are only available in a small fraction of centres in SSA. To overcome these limitations, many of the centres require larger numbers of experienced pathologists and technical staff. Furthermore, equipment maintenance, standardization of processing guidelines and establishment of an IHC service are needed to comply with international standards of breast cancer pathology.

Identification of Gastritis Subtypes by Convolutional Neuronal Networks on Histological Images of Antrum and Corpus Biopsies

Background: Gastritis is a prevalent disease and commonly classified into autoimmune (A), bacterial (B), and chemical (C) type gastritis. While the former two subtypes are associated with an increased risk of developing gastric intestinal adenocarcinoma, the latter subtype is not. In this study, we evaluated the capability to classify common gastritis subtypes using convolutional neuronal networks on a small dataset of antrum and corpus biopsies. Methods: 1230 representative 500 × 500 µm images of 135 patients with type A, type B, and type C gastritis were extracted from scanned histological slides. Patients were allocated randomly into a training set (60%), a validation set (20%), and a test set (20%). One classifier for antrum and one classifier for corpus were trained and optimized. After optimization, the test set was analyzed using a joint result from both classifiers. Results: Overall accuracy in the test set was 84% and was particularly high for type B gastritis with a sensitivity of 100% and a specificity of 93%. Conclusions: Classification of gastritis subtypes is possible using convolutional neural networks on a small dataset of histopathological images of antrum and corpus biopsies. Deep learning strategies to support routine diagnostic pathology merit further evaluation.