• machine learning
• proteome informatics
• cancer

• Carcinoma, Hepatocellular/diagnosis
• Carcinoma, Hepatocellular/pathology
• Liver Neoplasms/diagnosis
• Liver Neoplasms/pathology
• Humans
• Deep Learning
• Proteomics/methods
• Biomarkers, Tumor/analysis

• No.2022SCP002 Spring City Plan{:}the High-level Talent Promotion and Training Project of Kunming
• No. 2022ZFJH003 Fundamental Research Funds for the Central Universities
• No. 2019YFC0840600 and No. 2019YFC0840609 National Key Research and Development Program of China

• Analytical chemistry
• Artificial intelligence
• Chromatography
• Machine learning
• Spectroscopy

• artificial intelligence
• cholangiocarcinoma (CCA)
• diagnosis
• machine learning (ML)
• prognosis

• intraoperative assessment
• oral cancer
• resection margin
• spectroscopy

• ambient ionization
• clinical diagnostics
• handheld mass spectrometry–based devices
• mass spectrometry imaging

• Citespace
• VOSviewer
• artificial intelligence
• bibliometrics
• liver cancer

• Humans
• Spectrometry, Mass, Electrospray Ionization/methods
• Liver Neoplasms/diagnosis
• Carcinoma, Hepatocellular/diagnosis
• Machine Learning

In this paper, in-depth research analysis of anti-hepatocellular carcinoma molecular targets for hepatocellular carcinoma diagnosis was conducted using artificial intelligence. Because BRD4 plays an important role in gene transcription for cell cycle regulation and apoptosis, tumor-targeted therapy by inhibiting the expression or function of BRD4 has received increasing attention in the field of antitumor research. Study subjects in small samples were used as the validation set for validating each diagnostic model constructed based on the training set. The diagnostic effect of each model in the validation set is evaluated by calculating the sensitivity, specificity, and compliance rate, and the model with the best and most stable diagnostic value is selected by combining the results of model construction, validation, and evaluation. The total sample was divided into a training set and test set by using a stratified sampling method in the ratio of 7 : 3. Logistic regression, weighted k-nearest neighbor, decision tree, and BP artificial neural network were used in the training set to construct diagnostic models for early-stage liver cancer, respectively, and the optimal parameters of the corresponding models were obtained, and then, the constructed models were validated in the test set. To evaluate the diagnostic efficacy, stability, and generalization ability of the four classification methods more robustly, a 10-fold crossover test was performed for each classification method. BRD4 is an epigenetic regulator that is associated with the upregulation of expression of various oncogenic drivers in tumors. Targeting BRD4 with pharmacological inhibitors has emerged as a novel approach for tumor treatment. However, before we implemented this topic, there were no detailed studies on whether BRD4 could be used for the treatment of HCC, the role of BRD4 in HCC cell proliferation and apoptosis, and the ability of small molecule BRD4 inhibitors to induce apoptosis in hepatocellular carcinoma cells.

• machine learning
• mass spectrometry imaging
• papillary thyroid carcinoma
• probe electrospray ionization mass
• rapid diagnosis

• Humans
• Machine Learning
• Spectrometry, Mass, Electrospray Ionization
• Thyroid Cancer, Papillary/diagnosis
• Thyroid Neoplasms/diagnosis
• Thyroid Neoplasms/pathology
• Thyroid Nodule/pathology

Given the breakthroughs in key technologies, such as image recognition, deep learning and neural networks, artificial intelligence (AI) continues to be increasingly developed, leading to closer and deeper integration with an increasingly data-, knowledge- and brain labor-intensive medical industry. As society continues to advance and individuals become more aware of their health needs, the problems associated with the aging of the population are receiving increasing attention, and there is an urgent demand for improving medical technology, prolonging human life and enhancing health. Digestive system diseases are the most common clinical diseases and are characterized by complex clinical manifestations and a general lack of obvious symptoms in the early stage. Such diseases are very difficult to diagnose and treat. In recent years, the incidence of diseases of the digestive system has increased. As AI applications in the field of health care continue to be developed, AI has begun playing an important role in the diagnosis and treatment of diseases of the digestive system. In this paper, the application of AI in assisted diagnosis and the application and prospects of AI in malignant and benign digestive system diseases are reviewed.

• Artificial intelligence
• Digestive disease
• Convolutional neural network
• Deep learning
• Review

• Artificial Intelligence
• Deep Learning
• Gastrointestinal Diseases/diagnosis
• Gastrointestinal Diseases/therapy
• Humans
• Neural Networks, Computer

• Artificial intelligence
• Bioprinting
• Hepatocellular carcinoma
• Liver cancer
• Machine learning
• Three-dimensional printing

Rapid intraoperative diagnosis for unconfirmed pulmonary tumor is extremely important for determining the optimal surgical procedure (lobectomy or sublobar resection). Attempts to diagnose malignant tumors using mass spectrometry (MS) have recently been described. This study evaluated the usefulness of MS and artificial intelligence (AI) for differentiating primary lung adenocarcinoma (PLAC) and colorectal metastatic pulmonary tumor.

Pulmonary samples from 40 patients who underwent pulmonary resection for PLAC (20 tumors, 20 normal lungs) or pulmonary metastases originating from colorectal metastatic pulmonary tumor (CRMPT) (20 tumors, 20 normal lungs) were collected and analyzed retrospectively by probe electrospray ionization‐MS. AI using random forest (RF) algorithms was employed to evaluate the accuracy of each combination.

The accuracy of the machine learning algorithm applied using RF to distinguish malignant tumor (PLAC or CRMPT) from normal lung was 100%. The algorithms offered 97.2% accuracy in differentiating PLAC and CRMPT.

MS combined with an AI system demonstrated high accuracy not only for differentiating cancer from normal tissue, but also for differentiating between PLAC and CRMPT with a short working time. This method shows potential for application as a support tool facilitating rapid intraoperative diagnosis to determine the surgical procedure for pulmonary resection.

• mass spectrometry
• metastatic pulmonary tumor
• pulmonary adenocarcinoma

• deep learning
• diagnostic imaging
• liver fibrosis
• neural networks (computer)

• ambient mass spectrometry
• cancer diagnosis with ambient mass spectrometry
• lipid profiling
• untargeted lipidomics
• untargeted metabolomics

• Chronic liver disease
• Hepatologists
• LI-RADS
• Liver cancer
• Oncologists
• Radiology
• Surgeons

Preoperative histological evaluation of pancreatic neoplasms is important for guiding the resection strategy and preventing postoperative adverse events. However, conventional endoscopic methods have technical limitations that reduce the accuracy of the histopathological examination. Probe electrospray ionization mass spectrometry (PESI‐MS) may be a useful technique for rapidly evaluating small specimens.

This single‐center prospective study included patients with pancreatic neoplasms between October 2018 and December 2019. Pancreatic ductal adenocarcinoma (PDAC) specimens were obtained via endoscopic ultrasound‐guided fine needle aspiration (EUS‐FNA) and non‐neoplastic tissue was obtained via surgery. Specimens were subjected to PESI‐MS and the mass spectra were analyzed using partial least squares regression‐discriminant analysis.

The study included 40 patients with 20 nonneoplastic specimens and 19 PDAC specimens (1 case of neuroendocrine carcinoma was omitted). All nonneoplastic specimens were sufficient for PESI‐MS analysis, although only 7 of 19 PDAC specimens were sufficient for PESI‐MS analysis because of poor sample quality or insufficient quantity (<1 mg). Among the 27 analyzed cases, the mass spectra clearly differentiated between the PDAC and nonneoplastic specimens.

This study revealed that PESI‐MS could differentiate between PDAC and nonneoplastic specimens, even in cases where EUS‐FNA produced very small specimens.

• endoscopic ultrasound‐guided fine needle aspiration
• pancreatic ductal adenocarcinoma
• probe electrospray ionization mass spectrometry

• Lung cytopathology
• Mass spectrometry
• Probe electrospray ionization mass spectrometry
• Rapid cytological diagnosis

• Cell Culture Techniques/methods
• Cells, Cultured/pathology
• Cytodiagnosis/methods
• Humans
• Lung Neoplasms/pathology
• Mass Spectrometry/methods
• Spectrometry, Mass, Electrospray Ionization/methods

• Ambient ionization
• atmospheric pressure ionization
• high throughput analysis
• mass spectrometry

Hepatocellular carcinoma (HCC) is the most common primary cancer of the liver with high morbidity and mortality rates worldwide. Since 1963, when alpha-fetoprotein (AFP) was discovered as a first HCC serum biomarker, several other protein biomarkers have been identified and introduced into clinical practice. However, insufficient specificity and sensitivity of these biomarkers dictate the necessity of novel biomarker discovery. Remarkable advancements in integrated multiomics technologies for the identification of gene expression and protein or metabolite distribution patterns can facilitate rising to this challenge. Current multiomics technologies lead to the accumulation of a huge amount of data, which requires clustering and finding correlations between various datasets and developing predictive models for data filtering, pre-processing, and reducing dimensionality. Artificial intelligence (AI) technologies have an enormous potential to overcome accelerated data growth, complexity, and heterogeneity within and across data sources. Our review focuses on the recent progress in integrative proteomic profiling strategies and their usage in combination with machine learning and deep learning technologies for the discovery of novel biomarker candidates for HCC early diagnosis and prognosis. We discuss conventional and promising proteomic biomarkers of HCC such as AFP, lens culinaris agglutinin (LCA)-reactive L3 glycoform of AFP (AFP-L3), des-gamma-carboxyprothrombin (DCP), osteopontin (OPN), glypican-3 (GPC3), dickkopf-1 (DKK1), midkine (MDK), and squamous cell carcinoma antigen (SCCA) and highlight their functional significance including the involvement in cell signaling such as Wnt/β-catenin, PI3K/Akt, integrin αvβ3/NF-κB/HIF-1α, JAK/STAT3 and MAPK/ERK-mediated pathways dysregulated in HCC. We show that currently available computational platforms for big data analysis and AI technologies can both enhance proteomic profiling and improve imaging techniques to enhance the translational application of proteomics data into precision medicine.

• artificial intelligence
• biomarkers
• hepatocellular carcinoma
• proteomics
• translational medicine