Figure - available from: BioMed Research International
This content is subject to copyright.
(a) Normal; (b) Grade 1, well differentiated; (c) Grade 2, moderately differentiated; (d) Grade 3, poorly differentiated.
Source publication
Pathological diagnosis is influenced by subjective factors such as the individual experience and knowledge of doctors. Therefore, it may be interpreted in different ways for the same symptoms. The appearance of digital pathology has created good foundation for objective diagnoses based on quantitative feature analysis. Recently, numerous studies ar...
Citations
... Currently, limited research integrating AI with pancreatic pathological diagnoses focuses on extracting nuclear features related to DNA content and chromatin distribution from ERCP cytological specimens and surgically resected histological specimens [57][58][59]. For instance, Song et al. developed and assessed an SVM model for automatically diagnosing and grading PDAC based on the morphologic features found on histology slides, achieving an accuracy of 94.38% in binary classification between PDAC and normal tissues [60]. This outcome suggests a tremendous potential for this model as a valuable supplement for the morphological evaluation of tumor biological characteristics. ...
Solid pancreatic lesions (SPLs) encompass a variety of benign and malignant diseases and accurate diagnosis is crucial for guiding appropriate treatment decisions. Endoscopic ultrasonography-guided fine-needle aspiration/biopsy (EUS-FNA/B) serves as a front-line diagnostic tool for pancreatic mass lesions and is widely used in clinical practice. Artificial intelligence (AI) is a mathematical technique that automates the learning and recognition of data patterns. Its strong self-learning ability and unbiased nature have led to its gradual adoption in the medical field. In this paper, we describe the fundamentals of AI and provide a summary of reports on AI in EUS-FNA/B to help endoscopists understand and realize its potential in improving pathological diagnosis and guiding targeted EUS-FNA/B. However, AI models have limitations and shortages that need to be addressed before clinical use. Furthermore, as most AI studies are retrospective, large-scale prospective clinical trials are necessary to evaluate their clinical usefulness accurately. Although AI in EUS-FNA/B is still in its infancy, the constant input of clinical data and the advancements in computer technology are expected to make computer-aided diagnosis and treatment more feasible.
... AI can assist in the diagnosis of PC by analyzing cytology and biochemical characteristics of FNA/FNB samples. Here, we summarized the application of AI in pathological examination (Table 6) [76,[163][164][165][166][167][168][169]. ...
... They subsequently used multiple classifiers to demonstrate the effectiveness of the designed method in the differential diagnostic of SCN and MCN [163]. Using a similar approach, Song also worked on diagnosing and grading PDAC [164]. Kriegsmann et al. used CNN to construct models for automatic localization and quantification of tissue categories in whole tissue slides, including pancreatic intraepithelial neoplasia and PDAC [165]. ...
Pancreatic cancer is the deadliest disease, with a five-year overall survival rate of just 11%. The pancreatic cancer patients diagnosed with early screening have a median overall survival of nearly ten years, compared with 1.5 years for those not diagnosed with early screening. Therefore, early diagnosis and early treatment of pancreatic cancer are particularly critical. However, as a rare disease, the general screening cost of pancreatic cancer is high, the accuracy of existing tumor markers is not enough, and the efficacy of treatment methods is not exact. In terms of early diagnosis, artificial intelligence technology can quickly locate high-risk groups through medical images, pathological examination, biomarkers, and other aspects, then screening pancreatic cancer lesions early. At the same time, the artificial intelligence algorithm can also be used to predict the survival time, recurrence risk, metastasis, and therapy response which could affect the prognosis. In addition, artificial intelligence is widely used in pancreatic cancer health records, estimating medical imaging parameters, developing computer-aided diagnosis systems, etc. Advances in AI applications for pancreatic cancer will require a concerted effort among clinicians, basic scientists, statisticians, and engineers. Although it has some limitations, it will play an essential role in overcoming pancreatic cancer in the foreseeable future due to its mighty computing power.
... The designation of PDAC grading and staging are determined based on American Joint Committee Cancer (AJCC) Staging Manual [67,68] that take both the histopathological grading (G) and TNM scores into consideration [69,70]. The histopathological gradings (G1 to G3) are assigned based on the levels of glandular differentiation and pattern of tumour growth in the neoplastic pancreatic stroma on haematoxylin and eosin-stained tissue sections [69]. ...
Pancreatic ductal adenocarcinoma (PDAC), a highly aggressive malignancy with a poor prognosis is usually detected at the advanced stage of the disease. The only US Food and Drug Administration-approved biomarker that is available for PDAC, CA 19-9, is most useful in monitoring treatment response among PDAC patients rather than for early detection. Moreover, when CA 19-9 is solely used for diagnostic purposes, it has only a recorded sensitivity of 79% and specificity of 82% in symptomatic individuals. Therefore, there is an urgent need to identify reliable biomarkers for diagnosis (specifically for the early diagnosis), ascertain prognosis as well as to monitor treatment response and tumour recurrence of PDAC. In recent years, proteomic technologies are growing exponentially at an accelerated rate for a wide range of applications in cancer research. In this review, we discussed the current status of biomarker research for PDAC using various proteomic technologies. This review will explore the potential perspective for understanding and identifying the unique alterations in protein expressions that could prove beneficial in discovering new robust biomarkers to detect PDAC at an early stage, ascertain prognosis of patients with the disease in addition to monitoring treatment response and tumour recurrence of patients.
... Change et al. (30) used paired pancreatic histopathological and immunofluorescence images to classify nuclei. Song et al. (31) proposed a model for automatically grading pancreatic adenocarcinoma using morphological features. They segmented a PDAC tissue image into the lumen, epithelial nuclei, and non-epithelial nuclei, and then extracted several morphological features from the epithelial cells and segmented lumen parts, achieving an accuracy of 94.38% in binary classification. ...
Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancer types worldwide, with the lowest 5-year survival rate among all kinds of cancers. Histopathology image analysis is considered a gold standard for PDAC detection and diagnosis. However, the manual diagnosis used in current clinical practice is a tedious and time-consuming task and diagnosis concordance can be low. With the development of digital imaging and machine learning, several scholars have proposed PDAC analysis approaches based on feature extraction methods that rely on field knowledge. However, feature-based classification methods are applicable only to a specific problem and lack versatility, so that the deep-learning method is becoming a vital alternative to feature extraction. This paper proposes the first deep convolutional neural network architecture for classifying and segmenting pancreatic histopathological images on a relatively large WSI dataset. Our automatic patch-level approach achieved 95.3% classification accuracy and the WSI-level approach achieved 100%. Additionally, we visualized the classification and segmentation outcomes of histopathological images to determine which areas of an image are more important for PDAC identification. Experimental results demonstrate that our proposed model can effectively diagnose PDAC using histopathological images, which illustrates the potential of this practical application.
... To the best of our knowledge, this work represents the first classifier to successfully predict pancreatic cancer patient survival using only information from histopathology images. Previous work in the computational pathology of pancreatic cancer has instead focused on identification of cancerous glands and tissue regions 32,33 . Moreover, our analysis is the first to extract prognostic information from human, whole-slide tissue images of pancreatic cancer. ...
Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest forms of cancer, with an average 5-year survival rate of only 8%. Within PDAC patients, however, there is a small subset of patients who survive >10 years. Deciphering underlying reasons behind prolonged survival could potentially provide new opportunities to treat PDAC; however, no genomic, transcriptomic, proteomic, or clinical signatures have been found to robustly separate this subset of patients. Digital pathology, in combination with machine learning, provides an opportunity to computationally search for tissue morphology patterns associated with disease outcomes. Here, we developed a computational framework to analyze whole-slide images (WSI) of PDAC patient tissue and identify tissue-morphology signatures for very long term surviving patients. Our results indicate that less tissue morphology heterogeneity is significantly linked to better patient survival and that the extra-tumoral space encodes prognostic information for survival. Based on information from morphological heterogeneity in the tumor and its adjacent area, we established a machine learning model with an AUC of 0.94. Our analysis workflow highlighted a quantitative visual-based tissue phenotype analysis that also allows direct interaction with pathology. This study demonstrates a pathway to accelerate the discovery of undetermined tissue morphology associated with pathogenesis states and prognosis and diagnosis of patients by utilizing new computational approaches.
... 40 Song and Lee obtained results demonstrating an accuracy of 94.38% in distinguishing between pancreatic adenocarcinomas and normal tissue and reported a staging classification accuracy of 77.03% using a support vector machine to analyze the morphologic features present on histology slides. 41 In the current study, we describe a unique diagnostic application of image cytometry. We used hierarchical Kmeans clustering to segment images of cell clusters taken from pancreatic FNA cytology slides into nuclear and subnuclear ROIs and then trained an MNN to distinguish between benign and malignant cells using extracted features that parallel those used by cytopathologists in diagnosis. ...
Background:
Fine-needle aspiration (FNA) biopsy is an accurate method for the diagnosis of solid pancreatic masses. However, a significant number of cases still pose a diagnostic challenge. The authors have attempted to design a computer model to aid in the diagnosis of these biopsies.
Methods:
Images were captured of cell clusters on ThinPrep slides from 75 pancreatic FNA cases (20 malignant, 24 benign, and 31 atypical). A K-means clustering algorithm was used to segment the cell clusters into separable regions of interest before extracting features similar to those used for cytomorphologic assessment. A multilayer perceptron neural network (MNN) was trained and then tested for its ability to distinguish benign from malignant cases.
Results:
A total of 277 images of cell clusters were obtained. K-means clustering identified 68,301 possible regions of interest overall. Features such as contour, perimeter, and area were found to be significantly different between malignant and benign images (P <.05). The MNN was 100% accurate for benign and malignant categories. The model's predictions from the atypical data set were 77% accurate.
Conclusions:
The results of the current study demonstrate that computer models can be used successfully to distinguish benign from malignant pancreatic cytology. The fact that the model can categorize atypical cases into benign or malignant with 77% accuracy highlights the great potential of this technology. Although further study is warranted to validate its clinical applications in pancreatic and perhaps other areas of cytology as well, the potential for improved patient outcomes using MNN for image analysis in pathology is significant. Cancer Cytopathol 2017. © 2017 American Cancer Society.
... Early stage detection in mice [60] (C-MET, CK20, CEA) + CTCs elevated in late stages [96] miR-17-5p in serum exosomes correlates with stage [128] Distinguish Grade Ⅰ/Ⅱ in humans [61] Prognosis Potential CTC positivity has prognostic value in locally advanced pancreatic cancer [81] Potential CK20 expression in CTC indicates shorter overall survival [94] Monitor treatment Potential CTC levels decrease during 5-FU therapy [91] Potential Drug sensitivity/ pharmacokinetics CT scans can predict drug transport [35] CTC apoptosis can be detected after 5-FU therapy [91] Demonstrated for breast cancer [111] Monitor recurrence Potential CTC positivity correlates with postoperative staging [94][95][96][97] pathways: KRAS, transforming growth factor (TGF)-beta, WNT, NOTCH, ROBO/SLIT signaling, G1/S transition, SWI-SNF, chromatin modification, DNA repair and RNA processing [31] . Four tumor subtypes were identified based on differential expression of transcription factors and downstream targets: Squamous, pancreatic progenitor, immunogenic, and aberrantly differentiated endocrine exocrine (ADEX) tumors. ...
... These data sets are then used to train a predictive model that distinguishes normal tissue from premalignant cancer lesions [60] . Similar techniques can be extended to accomplish classification of PDAC by grade using human tissue samples [61] . Diagnosis of PDAC was made based on three parts: Segmentation and feature extraction; model learning and validation; and diagnosis. ...
... Diagnosis of PDAC was made based on three parts: Segmentation and feature extraction; model learning and validation; and diagnosis. Training data measuring ducts, consisting of the lumen and epithelial nuclei, can distinguish normal human subjects and those with grade Ⅰ and grade Ⅱ PDAC with an accuracy of 94% [61] . Automated systems have been developed for making a differential diagnosis of rare lesions such as cystic neoplasms of the pancreas using human biopsy tissue [62] . ...
Pancreatic cancer (PC) is a leading cause of cancer-related death worldwide. Clinical symptoms typically present late when treatment options are limited and survival expectancy is very short. Metastatic mutations are heterogeneous and can accumulate up to twenty years before PC diagnosis. Given such genetic diversity, detecting and managing the complex states of disease progression may be limited to imaging modalities and markers present in circulation. Recent developments in digital pathology imaging show potential for early PC detection, making a differential diagnosis, and predicting treatment sensitivity leading to long-term survival in advanced stage patients. Despite large research efforts, the only serum marker currently approved for clinical use is CA 19-9. Utility of CA 19-9 has been shown to improve when it is used in combination with PC-specific markers. Efforts are being made to develop early-screening assays that can detect tumor-derived material, present in circulation, before metastasis takes a significant course. Detection of markers that identify circulating tumor cells and tumor-derived extracellular vesicles (EVs) in biofluid samples offers a promising non-invasive method for this purpose. Circulating tumor cells exhibit varying expression of epithelial and mesenchymal markers depending on the state of tumor differentiation. This offers a possibility for monitoring disease progression using minimally invasive procedures. EVs also offer the benefit of detecting molecular cargo of tumor origin and add the potential to detect circulating vesicle markers from tumors that lack invasive properties. This review integrates recent genetic insights of PC progression with developments in digital pathology and early detection of tumor-derived circulating material.
... Patients suffering from pancreatic cancer have a very poor prognosis, with only 1-4% of patients surviving for five years. Pancreatic ductal adenocarcinoma (PDAC), composing 70% of all cases of pancreatic cancer, is not only the most common form, but also has the worst prognosis, of all of the pancreatic cancer subtypes (7). Currently, improvements in the treatment of pancreatic cancer patients are limited. ...
Endocrine gland‑derived vascular endothelial growth factor (EG‑VEGF) is a newly cloned factor that selectively acts on the endothelium of endocrine gland cells. EG‑VEGF was previously identified as an important cytokine, involved in the modulation of apoptosis in pancreatic cancer cell lines. The present study examined the effects of EG‑VEGF proliferation and migration, in pancreatic cancer cells. To determine the potential for EG‑VEGF as a therapeutic target for pancreatic cancer, the expression of EG‑VEGF were measured in pancreatic cancer tissue, and the association between its expression and the clinicopathological characteristics of the pancreatic cancer patients was determined. The results of the present study suggest that EG‑VEGF may act as a novel tumor gene in pancreatic cancer. EG‑VEGF was rarely expressed in the normal pancreatic tissue, but was highly expressed in the pancreatic cancer tissue. These data suggest that EG‑VEGF may be a cancer‑specific, and possibly tissue‑specific, survival factor in the pancreas. In the Mia PaCa‑2 pancreatic cancer cell line, EG‑VEGF was shown to promote proliferation and cellular invasion, and modulate the phosphorylation of mitogen‑activated protein kinase, a modulator for the malignant phenotype.
Artificial intelligence (AI) and machine learning (ML) have profoundly changed the landscape of healthcare. In pathology, AI/ML technologies are now able to enhance diagnostic accuracy, increase screening accuracy, and improve workflow efficiency. Notably, six such devices have been approved by the FDA. However, challenges persist, such as ensuring high-quality image datasets, robust validation processes, and managing regulatory hurdles. In this review, we cover AI/ML developments in pathology across several areas of study, including the cancers of the thyroid, cervix, bladder pancreas, and lung. Future advancements are expected to integrate cytopathology with radiology and clinical data, creating more robust and effective AI/ML tools for pathologists.
Artificial intelligence (AI) has been used in various fields of day-to-day life and its
role in medicine is immense. Understanding of oncology has been improved with
the introduction of AI which helps in diagnosis, treatment planning, management,
prognosis, and follow-up. It also helps to identify high-risk groups who can be
subjected to timely screening for early detection of malignant conditions. It is
more important in pancreatic cancer as it is one of the major causes of cancerrelated
deaths worldwide and there are no specific early features (clinical and
radiological) for diagnosis. With improvement in imaging modalities (computed
tomography, magnetic resonance imaging, endoscopic ultrasound), most often
clinicians were being challenged with lesions that were difficult to diagnose with
human competence. AI has been used in various other branches of medicine to
differentiate such indeterminate lesions including the thyroid gland, breast, lungs,
liver, adrenal gland, kidney, etc. In the case of pancreatic cancer, the role of AI has
been explored and is still ongoing. This review article will focus on how AI can be
used to diagnose pancreatic cancer early or differentiate it from benign pancreatic
lesions, therefore, management can be planned at an earlier stage.
