Performance and limitations of a supervised deep learning approach for the histopathological Oxford Classification of glomeruli with IgA nephropathy

BACKGROUND AND OBJECTIVE

The Oxford Classification for IgA nephropathy is the most successful example of an evidence-based nephropathology classification system. The aim of our study was to replicate the glomerular components of Oxford scoring with an end-to-end deep learning pipeline that involves automatic glomerular segmentation followed by classification for mesangial hypercellularity (M), endocapillary hypercellularity (E), segmental sclerosis (S) and active crescents (C).

METHODS

A total number of 1056 periodic acid-Schiff (PAS) whole slide images (WSIs), coming from 386 kidney biopsies, were annotated. Several detection models for glomeruli, based on the Mask R-CNN architecture, were trained on 587 WSIs, validated on 161 WSIs, and tested on 127 WSIs. For the development of segmentation models, 20,529 glomeruli were annotated, of which 16,571 as training and 3958 as validation set. The test set of the segmentation module comprised of 2948 glomeruli. For the Oxford classification, 6206 expert-annotated glomeruli from 308 PAS WSIs were labelled for M, E, S, C and split into a training set of 4298 glomeruli from 207 WSIs, and a test set of 1908 glomeruli. We chose the best-performing models to construct an end-to-end pipeline, which we named MESCnn (MESC classification by neural network), for the glomerular Oxford classification of WSIs.

RESULTS

Instance segmentation yielded excellent results with an AP50 ranging between 78.2-80.1 % (79.4 ± 0.7 %) on the validation and 75.1-77.7 % (76.5 ± 0.9 %) on the test set. The aggregated Jaccard Index was between 73.4-75.9 % (75.0 ± 0.8 %) on the validation and 69.1-73.4 % (72.2 ± 1.4 %) on the test set. At granular glomerular level, Oxford Classification was best replicated for M with EfficientNetV2-L with a mean ROC-AUC of 90.2 % and a mean precision/recall area under the curve (PR-AUC) of 81.8 %, best for E with MobileNetV2 (ROC-AUC 94.7 %) and ResNet50 (PR-AUC 75.8 %), best for S with EfficientNetV2-M (mean ROC-AUC 92.7 %, mean PR-AUC 87.7 %), best for C with EfficientNetV2-L (ROC-AUC 92.3 %) and EfficientNetV2-S (PR-AUC 54.7 %). At biopsy-level, correlation between expert and deep learning labels fulfilled the demands of the Oxford Classification.

CONCLUSION

We designed an end-to-end pipeline for glomerular Oxford Classification on both a granular glomerular and an entire biopsy level. Both the glomerular segmentation and the classification modules are freely available for further development to the renal medicine community.

ARGO 2.0: a Hybrid NLP/ML Framework for Diagnosis Standardization

A relevant problem in medicine is the standardization of the diagnosis associated with a clinical case. Although diagnosis formulation is an intrinsically subjective and uncertain process, its standardization may take benefit from digital solutions automating the routines at the basis of such a decision. In this work, we propose ARGO 2.0: a framework for the development of decision support systems for diagnosis formulation. The framework can read free-text reports and store their clinically relevant information as personalized electronic Case Report Forms. A hybrid strategy, exploiting the synergy of Natural Language Processing and Machine Learning techniques, is used to automatically suggest a diagnosis in a standardized fashion. ARGO 2.0 has been designed to be template-independent and easily tailored to specific medical fields. We here demonstrate its feasibility in hemo lympho-pathology, by detailing its implementation, object of an ongoing validation campaign in a standing medical institute. ARGO 2.0 achieved an average Accuracy of 95.07%, an average precision of 94.85%, an average Recall of 96.31% and a F-Score of 95.32% onto the test set, outperforming both its embedded components, based on Natural Language Processing and Machine Learning.

Flu vaccination coverage in a large Italian teaching hospital: the example of the leaders

Key messages

Annual flu vaccination among healthcare workers (HCWs) is recommended to prevent influenza and to avoid its transmission especially to frail patients. This study shows the growing flu vaccination coverage rate in our teaching hospital and the effectiveness of the example given by the vaccinated leaders in increasing the coverage among all HCWs.

A six-years point prevalence survey of healthcare-associated infections in an Italian hospital

Key messages

Point prevalence from 2013 to 2018 seems to be stable. An accurate incidence survey is needed in order to identify the main risk factors of HAI and to realize more specific hospital programmes. Length of stay at the moment of detection is useful to meaningfully predict HAIs prevalence although the cause-and-effect relationship is still not clear.

Incidence study of surgical site infections in a large University Hospital in Rome, first results

Background

Surgical Site Infections (SSI) represent about a quarter of healthcare associated infections (HAI), increase the risk of mortality and hospital stay and extend healthcare costs. The aim of this work is to evaluate the incidence of SSI in patients undergoing surgical procedures of Hospital Units in the Fondazione Policlinico A. Gemelli IRCCS (FPG).

Methods

This descriptive observational monocentric study has a 30-months duration: the first phase was conducted between July 2018 and March 2019. Patients undergoing abdominal surgery procedures were enrolled. Active infection surveillance was performed, according to the ECDC, during the hospital stay and the surgical outpatient visits. We also performed a post-discharge surveillance (PDS) at 30 days (or 90 days if an implant has been inserted), through a telephone interview, made by a medical doctor resident in Public Health.

Results

During the first phase were enrolled 431 procedures in 2 Hospital Units, about emergency and hepatobiliary surgery. There were 53 cases of surgical site infection (SSI) highlighted in these procedures. Therefore, SSI’s were reported in 12.09% of the procedures. In particular, in 6.28% of cases, SSIs were diagnosed during the hospital stay. The results of this study showed statistically significant differences (p < 0.005) in incidence rate of emergency (10,70%) and hepatobiliary tract (13,17%) procedures. The average hospital stay was significant higher (p < 0.001) in patients who developed an SSI (17.27 days) than in patients without SSI (4.89 days).

Conclusions

These preliminary results show that the incidence rate of SSI for gastrointestinal procedures is similar to other European Countries (9.6%). The Hospital Health Management of FPG take care to the surveillance of hospital infections and monitor the correct application of the procedures, in line with the standards defined by Joint Commission International for guarantee high level in patients’ care.

Key messages

Performing a survey to measure the incidence of surgical site infections for surgery procedures is very important to define it and to improve surveillance systems of hospital infections in FPG. Active surveillance of surgical site infections may allow to enhance appropriate preventive measures.