Deep Learning-Based Classification of Hepatocellular Nodular Lesions on Whole-Slide Histopathologic Images

Comparison between two artificial intelligence models to discriminate cancerous cell nuclei based on confocal fluorescence imaging in hepatocellular carcinoma

BACKGROUND

Hepatocellular carcinoma (HCC) exhibits an exceptional intratumoral heterogeneity that might influence diagnosis and outcome. Advances in digital microscopy and artificial intelligence (AI) may improve the HCC identification of liver cancer cells.

AIM

Two AI algorithms were designed to perform computer-assisted discrimination of tumour from non-tumour nuclei in HCC.

METHODS

Healthy livers and HCCs from commercially available tissue arrays were stained with an antibody against proliferating cell nuclear antigen and DRAQ5 dye with high affinity for double-stranded DNA, acquired by confocal microscopy imaging and then used to design machine learning (ML) and deep learning (DL) algorithms.

RESULTS

Nuclei were segmented and then used to develop the Model 1 and Model 2 algorithms, using ML and DL respectively. Model 1 was trained with some texture nuclear features extracted using discrete wavelet transform and grey-level co-occurrence matrix. Model 2 was trained with the segmented images without any additional information. The comparative analysis of the models showed that DL was more effective than ML, achieving an average accuracy of 88 % in discriminating healthy from neoplastic nuclei in HCC samples.

CONCLUSION

Our research shows that AI techniques and nuclear fluorescent staining could be useful tools for automatically detecting HCC cells in liver tissues.

Use of artificial intelligence for liver diseases: A survey from the EASL congress 2024

Artificial intelligence (AI) methods enable humans to analyse large amounts of data, which would otherwise not be feasibly quantifiable. This is especially true for unstructured visual and textual data, which can contain invaluable insights into disease. The hepatology research landscape is complex and has generated large amounts of data to be mined. Many open questions can potentially be addressed with existing data through AI methods. However, the field of AI is sometimes obscured by hype cycles and imprecise terminologies. This can conceal the fact that numerous hepatology research groups already use AI methods in their scientific studies. In this review article, we aim to assess the contemporaneous use of AI methods in hepatology in Europe. To achieve this, we systematically surveyed all scientific contributions presented at the EASL Congress 2024. Out of 1,857 accepted abstracts (1,712 posters and 145 oral presentations), 6 presentations (∼4%) and 69 posters (∼4%) utilised AI methods. Of these, 55 posters were included in this review, while the others were excluded due to missing posters or incomplete methodologies. Finally, we summarise current academic trends in the use of AI methods and outline future directions, providing guidance for scientific stakeholders in the field of hepatology.

Machine learning based on alcohol drinking-gut microbiota-liver axis in predicting the occurrence of early-stage hepatocellular carcinoma

BACKGROUND

Alcohol drinking and gut microbiota are related to hepatocellular carcinoma (HCC), but the specific relationship between them remains unclear.

AIMS

We aimed to establish the alcohol drinking-gut microbiota-liver axis and develop machine learning (ML) models in predicting the occurrence of early-stage HCC.

METHODS

Two hundred sixty-nine patients with early-stage HCC and 278 controls were recruited. Alcohol drinking-gut microbiota-liver axis was established through the mediation/moderation effect analyses. Eight ML algorithms including Classification and Regression Tree (CART), Gradient Boosting Machine (GBM), K-Nearest Neighbor (KNN), Logistic Regression (LR), Neural Network (NN), Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost) were applied.

RESULTS

A total of 160 pairs of individuals were included for analyses. The mediation effects of Genus_Catenibacterium (P = 0.024), Genus_Tyzzerella_4 (P < 0.001), and Species_Tyzzerella_4 (P = 0.020) were discovered. The moderation effects of Family_Enterococcaceae (OR = 0.741, 95%CI:0.160-0.760, P = 0.017), Family_Leuconostocaceae (OR = 0.793, 95%CI:0.486-3.593, P = 0.010), Genus_Enterococcus (OR = 0.744, 95%CI:0.161-0.753, P = 0.017), Genus_Erysipelatoclostridium (OR = 0.693, 95%CI:0.062-0.672, P = 0.032), Genus_Lactobacillus (OR = 0.655, 95%CI:0.098-0.749, P = 0.011), Species_Enterococcus_faecium (OR = 0.692, 95%CI:0.061-0.673, P = 0.013), and Species_Lactobacillus (OR = 0.653, 95%CI:0.086-0.765, P = 0.014) were uncovered. The predictive power of eight ML models was satisfactory (AUCs:0.855-0.932). The XGBoost model had the best predictive ability (AUC = 0.932).

CONCLUSIONS

ML models based on the alcohol drinking-gut microbiota-liver axis are valuable in predicting the occurrence of early-stage HCC.

The role of artificial intelligence in the management of liver diseases

Universal neonatal hepatitis B virus (HBV) vaccination and the advent of direct-acting antivirals (DAA) against hepatitis C virus (HCV) have reshaped the epidemiology of chronic liver diseases. However, some aspects of the management of chronic liver diseases remain unresolved. Nucleotide analogs can achieve sustained HBV DNA suppression but rarely lead to a functional cure. Despite the high efficacy of DAAs, successful antiviral therapy does not eliminate the risk of hepatocellular carcinoma (HCC), highlighted the need for cost-effective identification of high-risk populations for HCC surveillance and tailored HCC treatment strategies for these populations. The accessibility of high-throughput genomic data has accelerated the development of precision medicine, and the emergence of artificial intelligence (AI) has led to a new era of precision medicine. AI can learn from complex, non-linear data and identify hidden patterns within real-world datasets. The combination of AI and multi-omics approaches can facilitate disease diagnosis, biomarker discovery, and the prediction of treatment efficacy and prognosis. AI algorithms have been implemented in various aspects, including non-invasive tests, predictive models, image diagnosis, and the interpretation of histopathology findings. AI can support clinicians in decision-making, alleviate clinical burdens, and curtail healthcare expenses. In this review, we introduce the fundamental concepts of machine learning and review the role of AI in the management of chronic liver diseases.

Machine learning-based characterization of the gut microbiome associated with the progression of primary biliary cholangitis to cirrhosis

BACKGROUND

Primary biliary cholangitis (PBC) is associated closely with the gut microbiota. This study aimed to explore the characteristics of the gut microbiota after the progress of PBC to cirrhosis.

METHOD

This study focuses on utilizing the 16S rRNA gene sequencing method to screen for differences in gut microbiota in PBC patients who progress to cirrhosis. Then, we divided the data into training and verification sets and used seven different machine learning (ML) models to validate them respectively, calculating and comparing the accuracy, F1 score, precision, and recall, and screening the dominant intestinal flora affecting PBC cirrhosis.

RESULT

PBC cirrhosis patients showed decreased diversity and richness of gut microbiota. Additionally, there are alterations in the composition of gut microbiota in PBC cirrhosis patients. The abundance of Faecalibacterium and Gemmiger bacteria significantly decreases, while the abundance of Veillonella and Streptococcus significantly increases. Furthermore, machine learning methods identify Streptococcus and Gemmiger as the predominant gut microbiota in PBC patients with cirrhosis, serving as non-invasive biomarkers (AUC = 0.902).

CONCLUSION

Our study revealed that PBC cirrhosis patients gut microbiota composition and function have significantly changed. Streptococcus and Gemmiger may become a non-invasive biomarker for predicting the progression of PBC progress to cirrhosis.

Multiparametric MRI based deep learning model for prediction of early recurrence of hepatocellular carcinoma after SR following TACE

BACKGROUND

Surgical resection (SR) following transarterial chemoembolization (TACE) is a promising treatment for unresectable hepatocellular carcinoma (uHCC). However, biomarkers for the prediction of postoperative recurrence are needed.

PURPOSE

To develop and validate a model combining deep learning (DL) and clinical data for early recurrence (ER) in uHCC patients after TACE.

METHODS

A total of 511 patients who received SR following TACE were assigned to derivation (n = 413) and validation (n = 98) cohorts. Deep learning features were taken from the largest tumor area in liver MRI. A nomogram using DL signatures and clinical data was made to forecast early recurrence risk in uHCC patients. Model performance was evaluated using area under the curve (AUC).

RESULTS

A total of 2278 subsequences and 31,346 slices multiparametric MRI including contrast-enhanced T1WI, T2WI and DWI were input in the DL model simultaneously. Multivariable analysis identified three independent predictors for the development of the nomogram: tumor number (hazard ratio [HR]:3.42, 95% confidence interval [CI]: 2.75-4.31, P = 0.003), microvascular invasion (HR: 9.21, 6.24-32.14; P < 0.001), and DL scores (HR: 17.46, 95% CI: 12.94-23.57, P < 0.001). The AUC of the nomogram was 0.872 and 0.862 in two cohorts, significantly outperforming single-subsequence-based DL mode and clinical model (all, P < 0.001). The nomogram provided two risk strata for cumulative overall survival in two cohorts, showing significant statistical results (P < 0.001).

CONCLUSIONS

The DL-based nomogram is essential to identify patients with uHCC suitable for treatment with SR following TACE and may potentially benefit personalized decision-making.

Deep Learning Classification and Quantification of Pejorative and Nonpejorative Architectures in Resected Hepatocellular Carcinoma from Digital Histopathologic Images

Liver resection is one of the best treatments for small hepatocellular carcinoma (HCC), but post-resection recurrence is frequent. Biotherapies have emerged as an efficient adjuvant treatment, making the identification of patients at high risk of recurrence critical. Microvascular invasion (mVI), poor differentiation, pejorative macrotrabecular architectures, and vessels encapsulating tumor clusters architectures are the most accurate histologic predictors of recurrence, but their evaluation is time-consuming and imperfect. Herein, a supervised deep learning-based approach with ResNet34 on 680 whole slide images (WSIs) from 107 liver resection specimens was used to build an algorithm for the identification and quantification of these pejorative architectures. This model achieved an accuracy of 0.864 at patch level and 0.823 at WSI level. To assess its robustness, it was validated on an external cohort of 29 HCCs from another hospital, with an accuracy of 0.787 at WSI level, affirming its generalization capabilities. Moreover, the largest connected areas of the pejorative architectures extracted from the model were positively correlated to the presence of mVI and the number of tumor emboli. These results suggest that the identification of pejorative architectures could be an efficient surrogate of mVI and have a strong predictive value for the risk of recurrence. This study is the first step in the construction of a composite predictive algorithm for early post-resection recurrence of HCC, including artificial intelligence-based features.

CDK4/6 inhibitor PD-0332991 suppresses hepatocarcinogenesis by inducing senescence of hepatic tumor-initiating cells

INTRODUCTION

Owing to the limited treatment options for hepatocellular carcinoma (HCC), interventions targeting pre-HCC stages have attracted increasing attention. In the pre-HCC stage, hepatic tumor-initiating cells (hTICs) proliferate abnormally and contribute to hepatocarcinogenesis. Numerous studies have investigated targeted senescence induction as an HCC intervention. However, it remains to be clarified whether senescence induction of hTICs could serve as a pre-HCC intervention.

OBJECTIVES

This study was designed to investigate whether senescence induction of hTICs in the precancerous stage inhibit HCC initiation.

METHODS AND RESULTS

HCC models developed from chronic liver injury (CLI) were established by using Fah-/- mice and N-Ras + AKT mice. PD-0332991, a selective CDK4/6 inhibitor that blocks the G1/S transition in proliferating cells, was used to induce senescence during the pre-HCC stage. Upon administration of PD-0332991, we observed a significant reduction in HCC incidence following selective senescence induction in hTICs, and an alleviation liver injury in the CLI-HCC models. PD-0332991 also induced senescence in vitro in cultured hTICs isolated from CLI-HCC models. Moreover, RNA sequencing (RNA-seq) analysis delineated that the "Cyclin D-CDK4/6-INK4-Rb" pathway was activated in both mouse and human liver samples during the pre-HCC stage, while PD-0332991 exhibited substantial inhibition of this pathway, thereby inducing cellular senescence in hTICs. Regarding the immune microenvironment, we demonstrated that senescent hTICs secrete key senescence-associated secretory phenotypic (SASP) factors, CXCL10 and CCL2, to activate and recruit macrophages, and contribute to immune surveillance.

CONCLUSION

We found that hTICs can be targeted and induced into a senescent state during the pre-HCC stage. The SASP factors released by senescent hTICs further activate the immune response, facilitating the clearance of hTICs, and consequently suppressing HCC occurrence. We highlight the importance of pre-HCC interventions and propose that senescence-inducing drugs hold promise for preventing HCC initiation under CLI.

A graph-learning based model for automatic diagnosis of Sjögren's syndrome on digital pathological images: a multicentre cohort study

BACKGROUND

Sjögren's Syndrome (SS) is a rare chronic autoimmune disorder primarily affecting adult females, characterized by chronic inflammation and salivary and lacrimal gland dysfunction. It is often associated with systemic lupus erythematosus, rheumatoid arthritis and kidney disease, which can lead to increased mortality. Early diagnosis is critical, but traditional methods for diagnosing SS, mainly through histopathological evaluation of salivary gland tissue, have limitations.

METHODS

The study used 100 labial gland biopsy, creating whole-slide images (WSIs) for analysis. The proposed model, named Cell-tissue-graph-based pathological image analysis model (CTG-PAM) and based on graph theory, characterizes single-cell feature, cell-cell feature, and cell-tissue feature. Building upon these features, CTG-PAM achieves cellular-level classification, enabling lymphocyte recognition. Furthermore, it leverages connected component analysis techniques in the cell graph structure to perform SS diagnosis based on lymphocyte counts.

FINDINGS

CTG-PAM outperforms traditional deep learning methods in diagnosing SS. Its area under the receiver operating characteristic curve (AUC) is 1.0 for the internal validation dataset and 0.8035 for the external test dataset. This indicates high accuracy. The sensitivity of CTG-PAM for the external dataset is 98.21%, while the accuracy is 93.75%. In comparison, the sensitivity and accuracy for traditional deep learning methods (ResNet-50) are lower. The study also shows that CTG-PAM's diagnostic accuracy is closer to skilled pathologists compared to beginners.

INTERPRETATION

Our findings indicate that CTG-PAM is a reliable method for diagnosing SS. Additionally, CTG-PAM shows promise in enhancing the prognosis of SS patients and holds significant potential for the differential diagnosis of both non-neoplastic and neoplastic diseases. The AI model potentially extends its application to diagnosing immune cells in tumor microenvironments.

Artificial intelligence in liver cancer - new tools for research and patient management

Liver cancer has high incidence and mortality globally. Artificial intelligence (AI) has advanced rapidly, influencing cancer care. AI systems are already approved for clinical use in some tumour types (for example, colorectal cancer screening). Crucially, research demonstrates that AI can analyse histopathology, radiology and natural language in liver cancer, and can replace manual tasks and access hidden information in routinely available clinical data. However, for liver cancer, few of these applications have translated into large-scale clinical trials or clinically approved products. Here, we advocate for the incorporation of AI in all stages of liver cancer management. We present a taxonomy of AI approaches in liver cancer, highlighting areas with academic and commercial potential, and outline a policy for AI-based liver cancer management, including interdisciplinary training of researchers, clinicians and patients. The potential of AI in liver cancer is immense, but effort is required to ensure that AI can fulfil expectations.

Ultrasound-based deep learning radiomics nomogram for differentiating mass mastitis from invasive breast cancer

BACKGROUND

The purpose of this study is to develop and validate the potential value of the deep learning radiomics nomogram (DLRN) based on ultrasound to differentiate mass mastitis (MM) and invasive breast cancer (IBC).

METHODS

50 cases of MM and 180 cases of IBC with ultrasound Breast Imaging Reporting and Data System 4 category were recruited (training cohort, n = 161, validation cohort, n = 69). Based on PyRadiomics and ResNet50 extractors, radiomics and deep learning features were extracted, respectively. Based on supervised machine learning methods such as logistic regression, random forest, and support vector machine, as well as unsupervised machine learning methods using K-means clustering analysis, the differences in features between MM and IBC were analyzed to develop DLRN. The performance of DLRN had been evaluated by receiver operating characteristic curve, calibration, and clinical practicality.

RESULTS

Supervised machine learning results showed that compared with radiomics models, especially random forest models, deep learning models were better at recognizing MM and IBC. The area under the curve (AUC) of the validation cohort was 0.84, the accuracy was 0.83, the sensitivity was 0.73, and the specificity was 0.83. Compared to radiomics or deep learning models, DLRN even further improved discrimination ability (AUC of 0.90 and 0.90, accuracy of 0.83 and 0.88 for training and validation cohorts), which had better clinical benefits and good calibratability. In addition, the information heterogeneity of deep learning features in MM and IBC was validated again through unsupervised machine learning clustering analysis, indicating that MM had a unique features phenotype.

CONCLUSION

The DLRN developed based on radiomics and deep learning features of ultrasound images has potential clinical value in effectively distinguishing between MM and IBC. DLRN breaks through visual limitations and quantifies more image information related to MM based on computers, further utilizing machine learning to effectively utilize this information for clinical decision-making. As DLRN becomes an autonomous screening system, it will improve the recognition rate of MM in grassroots hospitals and reduce the possibility of incorrect treatment and overtreatment.

Thinking like a pathologist: Morphologic approach to hepatobiliary tumors by ChatGPT

OBJECTIVES

This research aimed to evaluate the effectiveness of ChatGPT in accurately diagnosing hepatobiliary tumors using histopathologic images.

METHODS

The study compared the diagnostic accuracies of the GPT-4 model, providing the same set of images and 2 different input prompts. The first prompt, the morphologic approach, was designed to mimic pathologists' approach to analyzing tissue morphology. In contrast, the second prompt functioned without incorporating this morphologic analysis feature. Diagnostic accuracy and consistency were analyzed.

RESULTS

A total of 120 photomicrographs, composed of 60 images of each hepatobiliary tumor and nonneoplastic liver tissue, were used. The findings revealed that the morphologic approach significantly enhanced the diagnostic accuracy and consistency of the artificial intelligence (AI). This version was particularly more accurate in identifying hepatocellular carcinoma (mean accuracy: 62.0% vs 27.3%), bile duct adenoma (10.7% vs 3.3%), and cholangiocarcinoma (68.7% vs 16.0%), as well as in distinguishing nonneoplastic liver tissues (77.3% vs 37.5%) (Ps ≤ .01). It also demonstrated higher diagnostic consistency than the other model without a morphologic analysis (κ: 0.46 vs 0.27).

CONCLUSIONS

This research emphasizes the importance of incorporating pathologists' diagnostic approaches into AI to enhance accuracy and consistency in medical diagnostics. It mainly showcases the AI's histopathologic promise when replicating expert diagnostic processes.

Deep learning-enabled classification of kidney allograft rejection on whole slide histopathologic images

Background

Diagnosis of kidney transplant rejection currently relies on manual histopathological assessment, which is subjective and susceptible to inter-observer variability, leading to limited reproducibility. We aim to develop a deep learning system for automated assessment of whole-slide images (WSIs) from kidney allograft biopsies to enable detection and subtyping of rejection and to predict the prognosis of rejection.

Method

We collected H&E-stained WSIs of kidney allograft biopsies at 400x magnification from January 2015 to September 2023 at two hospitals. These biopsy specimens were classified as T cell-mediated rejection, antibody-mediated rejection, and other lesions based on the consensus reached by two experienced transplant pathologists. To achieve feature extraction, feature aggregation, and global classification, we employed multi-instance learning and common convolution neural networks (CNNs). The performance of the developed models was evaluated using various metrics, including confusion matrix, receiver operating characteristic curves, the area under the curve (AUC), classification map, heat map, and pathologist-machine confrontations.

Results

In total, 906 WSIs from 302 kidney allograft biopsies were included for analysis. The model based on multi-instance learning enables detection and subtyping of rejection, named renal rejection artificial intelligence model (RRAIM), with the overall 3-category AUC of 0.798 in the independent test set, which is superior to that of three transplant pathologists under nearly routine assessment conditions. Moreover, the prognosis models accurately predicted graft loss within 1 year following rejection and treatment response for rejection, achieving AUC of 0.936 and 0.756, respectively.

Conclusion

We first developed deep-learning models utilizing multi-instance learning for the detection and subtyping of rejection and prediction of rejection prognosis in kidney allograft biopsies. These models performed well and may be useful in assisting the pathological diagnosis.

Deep Learning Histology for Prediction of Lymph Node Metastases and Tumor Regression after Neoadjuvant FLOT Therapy of Gastroesophageal Adenocarcinoma

BACKGROUND

The aim of this study was to establish a deep learning prediction model for neoadjuvant FLOT chemotherapy response. The neural network utilized clinical data and visual information from whole-slide images (WSIs) of therapy-naïve gastroesophageal cancer biopsies.

METHODS

This study included 78 patients from the University Hospital of Cologne and 59 patients from the University Hospital of Heidelberg used as external validation.

RESULTS

After surgical resection, 33 patients from Cologne (42.3%) were ypN0 and 45 patients (57.7%) were ypN+, while 23 patients from Heidelberg (39.0%) were ypN0 and 36 patients (61.0%) were ypN+ (p = 0.695). The neural network had an accuracy of 92.1% to predict lymph node metastasis and the area under the curve (AUC) was 0.726. A total of 43 patients from Cologne (55.1%) had less than 50% residual vital tumor (RVT) compared to 34 patients from Heidelberg (57.6%, p = 0.955). The model was able to predict tumor regression with an error of ±14.1% and an AUC of 0.648.

CONCLUSIONS

This study demonstrates that visual features extracted by deep learning from therapy-naïve biopsies of gastroesophageal adenocarcinomas correlate with positive lymph nodes and tumor regression. The results will be confirmed in prospective studies to achieve early allocation of patients to the most promising treatment.

A deep learning algorithm to identify carotid plaques and assess their stability

Background

Carotid plaques are major risk factors for stroke. Carotid ultrasound can help to assess the risk and incidence rate of stroke. However, large-scale carotid artery screening is time-consuming and laborious, the diagnostic results inevitably involve the subjectivity of the diagnostician to a certain extent. Deep learning demonstrates the ability to solve the aforementioned challenges. Thus, we attempted to develop an automated algorithm to provide a more consistent and objective diagnostic method and to identify the presence and stability of carotid plaques using deep learning.

Methods

A total of 3,860 ultrasound images from 1,339 participants who underwent carotid plaque assessment between January 2021 and March 2023 at the Shanghai Eighth People's Hospital were divided into a 4:1 ratio for training and internal testing. The external test included 1,564 ultrasound images from 674 participants who underwent carotid plaque assessment between January 2022 and May 2023 at Xinhua Hospital affiliated with Dalian University. Deep learning algorithms, based on the fusion of a bilinear convolutional neural network with a residual neural network (BCNN-ResNet), were used for modeling to detect carotid plaques and assess plaque stability. We chose AUC as the main evaluation index, along with accuracy, sensitivity, and specificity as auxiliary evaluation indices.

Results

Modeling for detecting carotid plaques involved training and internal testing on 1,291 ultrasound images, with 617 images showing plaques and 674 without plaques. The external test comprised 470 ultrasound images, including 321 images with plaques and 149 without. Modeling for assessing plaque stability involved training and internal testing on 764 ultrasound images, consisting of 494 images with unstable plaques and 270 with stable plaques. The external test was composed of 279 ultrasound images, including 197 images with unstable plaques and 82 with stable plaques. For the task of identifying the presence of carotid plaques, our model achieved an AUC of 0.989 (95% CI: 0.840, 0.998) with a sensitivity of 93.2% and a specificity of 99.21% on the internal test. On the external test, the AUC was 0.951 (95% CI: 0.962, 0.939) with a sensitivity of 95.3% and a specificity of 82.24%. For the task of identifying the stability of carotid plaques, our model achieved an AUC of 0.896 (95% CI: 0.865, 0.922) on the internal test with a sensitivity of 81.63% and a specificity of 87.27%. On the external test, the AUC was 0.854 (95% CI: 0.889, 0.830) with a sensitivity of 68.52% and a specificity of 89.49%.

Conclusion

Deep learning using BCNN-ResNet algorithms based on routine ultrasound images could be useful for detecting carotid plaques and assessing plaque instability.

Prognostic significance of TNFRSF4 expression and development of a pathomics model to predict expression in hepatocellular carcinoma

Background

TNFRSF4 plays a significant role in cancer progression, especially in hepatocellular carcinoma (HCC). This study aims to investigate the prognostic value of TNFRSF4 expression in patients with HCC and to develop a predictive pathomics model for its expression.

Methods

A cohort of patients with HCC retrieved from the TCGA database was analyzed using RNA-seq analysis to determine TNFRSF4 expression and its impact on overall survival (OS). Additionally, hematoxylin-eosin staining analysis was performed to construct a pathomics model for predicting TNFRSF4 expression. Then, pathway enrichment analysis was conducted, immune checkpoint markers were investigated, and immune cell infiltration was examined to explore the underlying biological mechanism of the pathomics score.

Results

TNFRSF4 expression was significantly higher in tumor tissues than in normal tissues. TNFRSF4 expression also exhibited significant correlations with various clinical variables, including pathologic stage III/IV and R1/R2/RX residual tumor. Furthermore, elevated TNFRSF4 expression was associated with unfavorable OS. Interestingly, in the subgroup analysis, elevated TNFRSF4 expression was identified as a significant risk factor for OS in male patients. The newly developed pathomics model successfully predicted TNFRSF4 expression with good performance and revealed a significant association between high pathomics scores and worse OS. In male patients, high pathomics scores were also associated with a higher risk of mortality. Moreover, pathomics scores were also involved in specific hallmarks, immune-related characteristics, and apoptosis-related genes in HCC, such as epithelial-mesenchymal transition, Tregs, and BAX expression.

Conclusions

Our findings suggest that TNFRSF4 expression and the newly devised pathomics scores hold potential as prognostic markers for OS in patients with HCC. Additionally, gender influenced the association between these markers and patient outcomes.

The Evolving Role of Artificial Intelligence in Gastrointestinal Histopathology: An Update

Significant advances in artificial intelligence (AI) over the past decade potentially may lead to dramatic effects on clinical practice. Digitized histology represents an area ripe for AI implementation. We describe several current needs within the world of gastrointestinal histopathology, and outline, using currently studied models, how AI potentially can address them. We also highlight pitfalls as AI makes inroads into clinical practice.

Deep learning based digital pathology for predicting treatment response to first-line PD-1 blockade in advanced gastric cancer

BACKGROUND

Advanced unresectable gastric cancer (GC) patients were previously treated with chemotherapy alone as the first-line therapy. However, with the Food and Drug Administration's (FDA) 2022 approval of programmed cell death protein 1 (PD-1) inhibitor combined with chemotherapy as the first-li ne treatment for advanced unresectable GC, patients have significantly benefited. However, the significant costs and potential adverse effects necessitate precise patient selection. In recent years, the advent of deep learning (DL) has revolutionized the medical field, particularly in predicting tumor treatment responses. Our study utilizes DL to analyze pathological images, aiming to predict first-line PD-1 combined chemotherapy response for advanced-stage GC.

METHODS

In this multicenter retrospective analysis, Hematoxylin and Eosin (H&E)-stained slides were collected from advanced GC patients across four medical centers. Treatment response was evaluated according to iRECIST 1.1 criteria after a comprehensive first-line PD-1 immunotherapy combined with chemotherapy. Three DL models were employed in an ensemble approach to create the immune checkpoint inhibitors Response Score (ICIsRS) as a novel histopathological biomarker derived from Whole Slide Images (WSIs).

RESULTS

Analyzing 148,181 patches from 313 WSIs of 264 advanced GC patients, the ensemble model exhibited superior predictive accuracy, leading to the creation of ICIsNet. The model demonstrated robust performance across four testing datasets, achieving AUC values of 0.92, 0.95, 0.96, and 1 respectively. The boxplot, constructed from the ICIsRS, reveals statistically significant disparities between the well response and poor response (all p-values < = 0.001).

CONCLUSION

ICIsRS, a DL-derived biomarker from WSIs, effectively predicts advanced GC patients' responses to PD-1 combined chemotherapy, offering a novel approach for personalized treatment planning and allowing for more individualized and potentially effective treatment strategies based on a patient's unique response situations.

Battle of the biopsies: Role of tissue and liquid biopsy in hepatocellular carcinoma

The diagnosis and management of hepatocellular carcinoma (HCC) have improved significantly in recent years. With the introduction of immunotherapy-based combination therapy, there has been a notable expansion in treatment options for patients with unresectable HCC. Simultaneously, innovative molecular tests for early detection and management of HCC are emerging. This progress prompts a key question: as liquid biopsy techniques rise in prominence, will they replace traditional tissue biopsies, or will both techniques remain relevant? Given the ongoing challenges of early HCC detection, including issues with ultrasound sensitivity, accessibility, and patient adherence to surveillance, the evolution of diagnostic techniques is more relevant than ever. Furthermore, the accurate stratification of HCC is limited by the absence of reliable biomarkers which can predict response to therapies. While the advantages of molecular diagnostics are evident, their potential has not yet been fully harnessed, largely because tissue biopsies are not routinely performed for HCC. Liquid biopsies, analysing components such as circulating tumour cells, DNA, and extracellular vesicles, provide a promising alternative, though they are still associated with challenges related to sensitivity, cost, and accessibility. The early results from multi-analyte liquid biopsy panels are promising and suggest they could play a transformative role in HCC detection and management; however, comprehensive clinical validation is still ongoing. In this review, we explore the challenges and potential of both tissue and liquid biopsy, highlighting that these diagnostic methods, while distinct in their approaches, are set to jointly reshape the future of HCC management.

Primary liver cancer classification from routine tumour biopsy using weakly supervised deep learning

Background & Aims

The diagnosis of primary liver cancers (PLCs) can be challenging, especially on biopsies and for combined hepatocellular-cholangiocarcinoma (cHCC-CCA). We automatically classified PLCs on routine-stained biopsies using a weakly supervised learning method.

Method

We selected 166 PLC biopsies divided into training, internal and external validation sets: 90, 29 and 47 samples, respectively. Two liver pathologists reviewed each whole-slide hematein eosin saffron (HES)-stained image (WSI). After annotating the tumour/non-tumour areas, tiles of 256x256 pixels were extracted from the WSIs and used to train a ResNet18 neural network. The tumour/non-tumour annotations served as labels during training, and the network's last convolutional layer was used to extract new tumour tile features. Without knowledge of the precise labels of the malignancies, we then applied an unsupervised clustering algorithm.

Results

Pathological review classified the training and validation sets into hepatocellular carcinoma (HCC, 33/90, 11/29 and 26/47), intrahepatic cholangiocarcinoma (iCCA, 28/90, 9/29 and 15/47), and cHCC-CCA (29/90, 9/29 and 6/47). In the two-cluster model, Clusters 0 and 1 contained mainly HCC and iCCA histological features. The diagnostic agreement between the pathological diagnosis and the two-cluster model predictions (major contingent) in the internal and external validation sets was 100% (11/11) and 96% (25/26) for HCC and 78% (7/9) and 87% (13/15) for iCCA, respectively. For cHCC-CCA, we observed a highly variable proportion of tiles from each cluster (cluster 0: 5-97%; cluster 1: 2-94%).

Conclusion

Our method applied to PLC HES biopsy could identify specific morphological features of HCC and iCCA. Although no specific features of cHCC-CCA were recognized, assessing the proportion of HCC and iCCA tiles within a slide could facilitate the identification of cHCC-CCA.

Impact and implications

The diagnosis of primary liver cancers can be challenging, especially on biopsies and for combined hepatocellular-cholangiocarcinoma (cHCC-CCA). We automatically classified primary liver cancers on routine-stained biopsies using a weakly supervised learning method. Our model identified specific features of hepatocellular carcinoma and intrahepatic cholangiocarcinoma. Despite no specific features of cHCC-CCA being recognized, the identification of hepatocellular carcinoma and intrahepatic cholangiocarcinoma tiles within a slide could facilitate the diagnosis of primary liver cancers, and particularly cHCC-CCA.

Precision treatment in advanced hepatocellular carcinoma

The past decade has witnessed significant advances in the systemic treatment of advanced hepatocellular carcinoma (HCC). Nevertheless, the newly developed treatment strategies have not achieved universal success and HCC patients frequently exhibit therapeutic resistance to these therapies. Precision treatment represents a paradigm shift in cancer treatment in recent years. This approach utilizes the unique molecular characteristics of individual patient to personalize treatment modalities, aiming to maximize therapeutic efficacy while minimizing side effects. Although precision treatment has shown significant success in multiple cancer types, its application in HCC remains in its infancy. In this review, we discuss key aspects of precision treatment in HCC, including therapeutic biomarkers, molecular classifications, and the heterogeneity of the tumor microenvironment. We also propose future directions, ranging from revolutionizing current treatment methodologies to personalizing therapy through functional assays, which will accelerate the next phase of advancements in this area.

Chinese experts' consensus on the application of intensive care big data

The development of intensive care medicine is inseparable from the diversified monitoring data. Intensive care medicine has been closely integrated with data since its birth. Critical care research requires an integrative approach that embraces the complexity of critical illness and the computational technology and algorithms that can make it possible. Considering the need of standardization of application of big data in intensive care, Intensive Care Medicine Branch of China Health Information and Health Care Big Data Society, Standard Committee has convened expert group, secretary group and the external audit expert group to formulate Chinese Experts' Consensus on the Application of Intensive Care Big Data (2022). This consensus makes 29 recommendations on the following five parts: Concept of intensive care big data, Important scientific issues, Standards and principles of database, Methodology in solving big data problems, Clinical application and safety consideration of intensive care big data. The consensus group believes this consensus is the starting step of application big data in the field of intensive care. More explorations and big data based retrospective research should be carried out in order to enhance safety and reliability of big data based models of critical care field.

Identifying immune infiltration by deep learning to assess the prognosis of patients with hepatocellular carcinoma

BACKGROUND

The treatment situation for hepatocellular carcinoma remains critical. The use of deep learning algorithms to assess immune infiltration is a promising new diagnostic tool.

METHODS

Patient data and whole slide images (WSIs) were obtained for the Xijing Hospital (XJH) cohort and TCGA cohort. We wrote programs using Visual studio 2022 with C# language to segment the WSI into tiles. Pathologists classified the tiles and later trained deep learning models using the ResNet 101V2 network via ML.NET with the TensorFlow framework. Model performance was evaluated using AccuracyMicro versus AccuracyMacro. Model performance was examined using ROC curves versus PR curves. The percentage of immune infiltration was calculated using the R package survminer to calculate the intergroup cutoff, and the Kaplan‒Meier method was used to plot the overall survival curve of patients. Cox regression was used to determine whether the percentage of immune infiltration was an independent risk factor for prognosis. A nomogram was constructed, and its accuracy was verified using time-dependent ROC curves with calibration curves. The CIBERSORT algorithm was used to assess immune infiltration between groups. Gene Ontology was used to explore the pathways of differentially expressed genes.

RESULTS

There were 100 WSIs and 165,293 tiles in the training set. The final deep learning models had an AccuracyMicro of 97.46% and an AccuracyMacro of 82.28%. The AUCs of the ROC curves on both the training and validation sets exceeded 0.95. The areas under the classification PR curves exceeded 0.85, except that of the TLS on the validation set, which might have had poor results (0.713) due to too few samples. There was a significant difference in OS between the TIL classification groups (p < 0.001), while there was no significant difference in OS between the TLS groups (p = 0.294). Cox regression showed that TIL percentage was an independent risk factor for prognosis in HCC patients (p = 0.015). The AUCs according to the nomogram were 0.714, 0.690, and 0.676 for the 1-year, 2-year, and 5-year AUCs in the TCGA cohort and 0.756, 0.797, and 0.883 in the XJH cohort, respectively. There were significant differences in the levels of infiltration of seven immune cell types between the two groups of samples, and gene ontology showed that the differentially expressed genes between the groups were immune related. Their expression levels of PD-1 and CTLA4 were also significantly different.

CONCLUSION

We constructed and tested a deep learning model that evaluates the immune infiltration of liver cancer tissue in HCC patients. Our findings demonstrate the value of the model in assessing patient prognosis, immune infiltration and immune checkpoint expression levels.

Deep Learning-Enabled Diagnosis of Liver Adenocarcinoma

BACKGROUND & AIMS

Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision making. However, rendering a correct diagnosis can be challenging, and often requires the integration of clinical, radiologic, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma from colorectal liver metastasis, as the most frequent primary and secondary forms of liver adenocarcinoma, with clinical grade accuracy using H&E-stained whole-slide images.

METHODS

HEPNET was trained on 714,589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital.

RESULTS

On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve of 0.994 (95% CI, 0.989-1.000) and an accuracy of 96.522% (95% CI, 94.521%-98.694%) at the patient level. Validation on the external test set yielded an area under the receiver operating characteristic curve of 0.997 (95% CI, 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI, 96.907%-100.000%). HEPNET surpassed the performance of 6 pathology experts with different levels of experience in a reader study of 50 patients (P = .0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses.

CONCLUSIONS

We provided a ready-to-use tool with clinical grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings.

Real-time and accurate estimation of surgical hemoglobin loss using deep learning-based medical sponges image analysis

Real-time and accurate estimation of surgical hemoglobin (Hb) loss is essential for fluid resuscitation management and evaluation of surgical techniques. In this study, we aimed to explore a novel surgical Hb loss estimation method using deep learning-based medical sponges image analysis. Whole blood samples of pre-measured Hb concentration were collected, and normal saline was added to simulate varying levels of Hb concentration. These blood samples were distributed across blank medical sponges to generate blood-soaked sponges. Eight hundred fifty-one blood-soaked sponges representing a wide range of blood dilutions were randomly divided 7:3 into a training group (n = 595) and a testing group (n = 256). A deep learning model based on the YOLOv5 network was used as the target region extraction and detection, and the three models (Feature extraction technology, ResNet-50, and SE-ResNet50) were trained to predict surgical Hb loss. Mean absolute error (MAE), mean absolute percentage error (MAPE), coefficient (R2) value, and the Bland-Altman analysis were calculated to evaluate the predictive performance in the testing group. The deep learning model based on SE-ResNet50 could predict surgical Hb loss with the best performance (R2 = 0.99, MAE = 11.09 mg, MAPE = 8.6%) compared with other predictive models, and Bland-Altman analysis also showed a bias of 1.343 mg with narrow limits of agreement (- 29.81 to 32.5 mg) between predictive and actual Hb loss. The interactive interface was also designed to display the real-time prediction of surgical Hb loss more intuitively. Thus, it is feasible for real-time estimation of surgical Hb loss using deep learning-based medical sponges image analysis, which was helpful for clinical decisions and technical evaluation.

Efficient diagnosis of IDH-mutant gliomas: 1p/19qNET assesses 1p/19q codeletion status using weakly-supervised learning

Accurate identification of molecular alterations in gliomas is crucial for their diagnosis and treatment. Although, fluorescence in situ hybridization (FISH) allows for the observation of diverse and heterogeneous alterations, it is inherently time-consuming and challenging due to the limitations of the molecular method. Here, we report the development of 1p/19qNET, an advanced deep-learning network designed to predict fold change values of 1p and 19q chromosomes and classify isocitrate dehydrogenase (IDH)-mutant gliomas from whole-slide images. We trained 1p/19qNET on next-generation sequencing data from a discovery set (DS) of 288 patients and utilized a weakly-supervised approach with slide-level labels to reduce bias and workload. We then performed validation on an independent validation set (IVS) comprising 385 samples from The Cancer Genome Atlas, a comprehensive cancer genomics resource. 1p/19qNET outperformed traditional FISH, achieving R2 values of 0.589 and 0.547 for the 1p and 19q arms, respectively. As an IDH-mutant glioma classifier, 1p/19qNET attained AUCs of 0.930 and 0.837 in the DS and IVS, respectively. The weakly-supervised nature of 1p/19qNET provides explainable heatmaps for the results. This study demonstrates the successful use of deep learning for precise determination of 1p/19q codeletion status and classification of IDH-mutant gliomas as astrocytoma or oligodendroglioma. 1p/19qNET offers comparable results to FISH and provides informative spatial information. This approach has broader applications in tumor classification.

Discrimination of early HCC using single cell nucleus image and visualization of feature distribution in whole slide images

Among hepatocellular carcinoma (HCC), early HCC such as well-differentiated hepatocellular carcinoma is more difficult to distinguish from non-cancer than other cancers. In particular, very well-differentiated hepatocellular carcinoma is even more difficult to distinguish, and it is difficult for pathologists to distinguish between cancer and non-cancer from a single nucleus image. If a function to distinguish cancer with a single cell nucleus image is realized, it may be possible to find new features related to nuclei that are useful for differentiating early HCC. The function will also be very helpful in needle biopsy where the area that can be observed is limited.In this study, we investigated the potential to discriminate cancer/non-cancer from an image of a single hepatocyte nucleus using CNN. The results indicated that discrimination was achievable with a correct rate of around 70%.The probability of cancer/non-cancer was visualized on WSI. The visualization results indicated a difference between cancerous and non-cancerous areas in 71% of the cases, which will help pathologists distinguish region of interest. Grouping sections with similar features also proved useful in improving accuracy and visualization results.

Artificial Intelligence-Based Opportunities in Liver Pathology-A Systematic Review

BACKGROUND

Artificial Intelligence (AI)-based Deep Neural Networks (DNNs) can handle a wide range of applications in image analysis, ranging from automated segmentation to diagnostic and prediction. As such, they have revolutionized healthcare, including in the liver pathology field.

OBJECTIVE

The present study aims to provide a systematic review of applications and performances provided by DNN algorithms in liver pathology throughout the Pubmed and Embase databases up to December 2022, for tumoral, metabolic and inflammatory fields.

RESULTS

42 articles were selected and fully reviewed. Each article was evaluated through the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool, highlighting their risks of bias.

CONCLUSIONS

DNN-based models are well represented in the field of liver pathology, and their applications are diverse. Most studies, however, presented at least one domain with a high risk of bias according to the QUADAS-2 tool. Hence, DNN models in liver pathology present future opportunities and persistent limitations. To our knowledge, this review is the first one solely focused on DNN-based applications in liver pathology, and to evaluate their bias through the lens of the QUADAS2 tool.

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.

Predicting the recurrence and overall survival of patients with glioma based on histopathological images using deep learning

Background

A deep learning (DL) model based on representative biopsy tissues can predict the recurrence and overall survival of patients with glioma, leading to optimized personalized medicine. This research aimed to develop a DL model based on hematoxylin-eosin (HE) stained pathological images and verify its diagnostic accuracy.

Methods

Our study retrospectively collected 162 patients with glioma and randomly divided them into a training set (n = 113) and a validation set (n = 49) to build a DL model. The HE-stained slide was segmented into a size of 180 × 180 pixels without overlapping. The patch-level features were extracted by the pre-trained ResNet50 to predict the recurrence and overall survival. Additionally, a light-strategy was introduced where low-size digital biopsy images with clinical information were inputted into the DL model to ensure minimum memory occupation.

Results

Our study extracted 512 histopathological features from the HE-stained slides of each glioma patient. We identified 36 and 18 features as significantly related to disease-free survival (DFS) and overall survival (OS), respectively, (P &lt; 0.05) using the univariate Cox proportional-hazards model. Pathomics signature showed a C-index of 0.630 and 0.652 for DFS and OS prediction, respectively. The time-dependent receiver operating characteristic (ROC) curves, along with nomograms, were used to assess the diagnostic accuracy at a fixed time point. In the validation set (n = 49), the area under the curve (AUC) in the 1- and 2-year DFS was 0.955 and 0.904, respectively, and the 2-, 3-, and 5-year OS were 0.969, 0.955, and 0.960, respectively. We stratified the patients into low- and high-risk groups using the median hazard score (0.083 for DFS and−0.177 for OS) and showed significant differences between these groups (P &lt; 0.001).

Conclusion

Our results demonstrated that the DL model based on the HE-stained slides showed the predictability of recurrence and survival in patients with glioma. The results can be used to assist oncologists in selecting the optimal treatment strategy in clinical practice.

An artificial intelligence model for the pathological diagnosis of invasion depth and histologic grade in bladder cancer

BACKGROUND

Accurate pathological diagnosis of invasion depth and histologic grade is key for clinical management in patients with bladder cancer (BCa), but it is labour-intensive, experience-dependent and subject to interobserver variability. Here, we aimed to develop a pathological artificial intelligence diagnostic model (PAIDM) for BCa diagnosis.

METHODS

A total of 854 whole slide images (WSIs) from 692 patients were included and divided into training and validation sets. The PAIDM was developed using the training set based on the deep learning algorithm ScanNet, and the performance was verified at the patch level in validation set 1 and at the WSI level in validation set 2. An independent validation cohort (validation set 3) was employed to compare the PAIDM and pathologists. Model performance was evaluated using the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value and negative predictive value.

RESULTS

The AUCs of the PAIDM were 0.878 (95% CI 0.875-0.881) at the patch level in validation set 1 and 0.870 (95% CI 0.805-0.923) at the WSI level in validation set 2. In comparing the PAIDM and pathologists, the PAIDM achieved an AUC of 0.847 (95% CI 0.779-0.905), which was non-inferior to the average diagnostic level of pathologists. There was high consistency between the model-predicted and manually annotated areas, improving the PAIDM's interpretability.

CONCLUSIONS

We reported an artificial intelligence-based diagnostic model for BCa that performed well in identifying invasion depth and histologic grade. Importantly, the PAIDM performed admirably in patch-level recognition, with a promising application for transurethral resection specimens.

Prognostic Value of CD8+ Lymphocytes in Hepatocellular Carcinoma and Perineoplastic Parenchyma Assessed by Interface Density Profiles in Liver Resection Samples

Hepatocellular carcinoma (HCC) often emerges in the setting of long-standing inflammatory liver disease. CD8 lymphocytes are involved in both the antitumoral response and hepatocyte damage in the remaining parenchyma. We investigated the dual role of CD8 lymphocytes by assessing density profiles at the interfaces of both HCC and perineoplastic liver parenchyma with surrounding stroma in whole-slide immunohistochemistry images of surgical resection samples. We applied a hexagonal grid-based digital image analysis method to sample the interface zones and compute the CD8 density profiles within them. The prognostic value of the indicators was explored in the context of clinicopathological, peripheral blood testing, and surgery data. Independent predictors of worse OS were a low standard deviation of CD8+ density along the tumor edge, high mean CD8+ density within the epithelial aspect of the perineoplastic liver-stroma interface, longer duration of surgery, a higher level of aspartate transaminase (AST), and a higher basophil count in the peripheral blood. A combined score, derived from these five independent predictors, enabled risk stratification of the patients into three prognostic categories with a 5-year OS probability of 76%, 40%, and 8%. Independent predictors of longer RFS were stage pT1, shorter duration of surgery, larger tumor size, wider tumor-free margin, and higher mean CD8+ density in the epithelial aspect of the tumor-stroma interface. We conclude that (1) our computational models reveal independent and opposite prognostic impacts of CD8+ cell densities at the interfaces of the malignant and non-malignant epithelium interfaces with the surrounding stroma; and (2) together with pathology, surgery, and laboratory data, comprehensive prognostic models can be constructed to predict patient outcomes after liver resection due to HCC.

The potential role of machine learning in modelling advanced chronic liver disease

The use of artificial intelligence is rapidly increasing in medicine to support clinical decision making mostly through diagnostic and prediction models. Such models derive from huge databases (big data) including a large variety of health-related individual patient data (input) and the corresponding diagnosis and/or outcome (labels). Various types of algorithms (e.g. neural networks) based on powerful computational ability (machine), allow to detect the relationship between input and labels (learning). More complex algorithms, like recurrent neural network can learn from previous as well as actual input (deep learning) and are used for more complex tasks like imaging analysis and personalized (bespoke) medicine. The prompt availability of big data makes that artificial intelligence can provide rapid answers to questions that would require years of traditional clinical research. It may therefore be a key tool to overcome several major gaps in the model of advanced chronic liver disease, mostly transition from mild to clinically significant portal hypertension, the impact of acute decompensation and the role of further decompensation and treatment efficiency. However, several limitations of artificial intelligence should be overcome before its application in clinical practice. Assessment of the risk of bias, understandability of the black boxes developing the models and models' validation are the most important areas deserving clarification for artificial intelligence to be widely accepted from physicians and patients.

Assessing electrocardiogram changes after ischemic stroke with artificial intelligence

OBJECTIVE

Ischemic stroke (IS) with subsequent cerebrocardiac syndrome (CCS) has a poor prognosis. We aimed to investigate electrocardiogram (ECG) changes after IS with artificial intelligence (AI).

METHODS

We collected ECGs from a healthy population and patients with IS, and then analyzed participant demographics and ECG parameters to identify abnormal features in post-IS ECGs. Next, we trained the convolutional neural network (CNN), random forest (RF) and support vector machine (SVM) models to automatically detect the changes in the ECGs; Additionally, We compared the CNN scores of good prognosis (mRS ≤ 2) and poor prognosis (mRS > 2) to assess the prognostic value of CNN model. Finally, we used gradient class activation map (Grad-CAM) to localize the key abnormalities.

RESULTS

Among the 3506 ECGs of the IS patients, 2764 ECGs (78.84%) led to an abnormal diagnosis. Then we divided ECGs in the primary cohort into three groups, normal ECGs (N-Ns), abnormal ECGs after the first ischemic stroke (A-ISs), and normal ECGs after the first ischemic stroke (N-ISs). Basic demographic and ECG parameter analyses showed that heart rate, QT interval, and P-R interval were significantly different between 673 N-ISs and 3546 N-Ns (p < 0.05). The CNN has the best performance among the three models in distinguishing A-ISs and N-Ns (AUC: 0.88, 95%CI = 0.86-0.90). The prediction scores of the A-ISs and N-ISs obtained from the all three models are statistically different from the N-Ns (p < 0.001). Futhermore, the CNN scores of the two groups (mRS > 2 and mRS ≤ 2) were significantly different (p < 0.05). Finally, Grad-CAM revealed that the V4 lead may harbor the highest probability of abnormality.

CONCLUSION

Our study showed that a high proportion of post-IS ECGs harbored abnormal changes. Our CNN model can systematically assess anomalies in and prognosticate post-IS ECGs.

Artificial intelligence for precision medicine in autoimmune liver disease

Autoimmune liver diseases (AiLDs) are rare autoimmune conditions of the liver and the biliary tree with unknown etiology and limited treatment options. AiLDs are inherently characterized by a high degree of complexity, which poses great challenges in understanding their etiopathogenesis, developing novel biomarkers and risk-stratification tools, and, eventually, generating new drugs. Artificial intelligence (AI) is considered one of the best candidates to support researchers and clinicians in making sense of biological complexity. In this review, we offer a primer on AI and machine learning for clinicians, and discuss recent available literature on its applications in medicine and more specifically how it can help to tackle major unmet needs in AiLDs.

Deep Learning-Based Prediction of Treatment Prognosis from Nasal Polyp Histology Slides

BACKGROUND

Histopathology of nasal polyps contains rich prognostic information, which is difficult to objectively extract. In the present study, we aimed to develop a prognostic indicator of patient outcomes by analyzing scanned conventional haematoxylin and eosin (H&E) -stained slides alone using deep learning.

METHODS

An interpretable supervised deep learning model was developed using 185 H&E-stained whole-slide images (WSIs) of nasal polyps, each from a patient randomly selected from the pool of 232 patients who underwent endoscopic sinus surgery at the First Affiliated Hospital of Sun Yat-sen University (internal cohort). We internally validated the model on a holdout dataset from the internal cohort (47 H&E-stained WSIs) and externally validated the model on 122 H&E-stained WSIs from the Seventh Affiliated Hospital of Sun Yat-sen University and the University of Hong Kong-Shenzhen Hospital (external cohort). A poor prognosis score (PPS) was established to evaluate patient outcomes, and then risk activation mapping was applied to visualize the histopathological features underlying PPS.

RESULTS

The model yielded a patient-level sensitivity of 79.5%, and specificity of 92.3%, with areas under the receiver operating characteristic curve of 0.943, on the multi-center external cohort. The predictive ability of PPS was superior to that of conventional tissue eosinophil number. Notably, eosinophil infiltration, goblet cell hyperplasia, glandular hyperplasia, squamous metaplasia, and fibrin deposition were identified as the main underlying features of PPS.

CONCLUSIONS

Our deep learning model is an effective method for decoding pathological images of nasal polyps, providing a valuable solution for disease prognosis prediction and precise patient treatment. This article is protected by copyright. All rights reserved.