Artificial intelligence-aided diagnostic imaging: A state-of-the-art technique in precancerous screening.

BACKGROUND AND AIM: Chromoendoscopy with the use of indigo carmine (IC) dye is a crucial endoscopic technique to identify gastrointestinal neoplasms. However, its performance is limited by the endoscopist's skill, and no standards are available for lesion identification. Thus, we developed an artificial intelligence (AI) model to replace chromoendoscopy. METHODS: This pilot study assessed the feasibility of our novel AI model in the conversion of white-light images (WLI) into virtual IC-dyed images based on a generative adversarial network. The predictions of our AI model were evaluated against the assessments of five endoscopic experts who were blinded to the purpose of this study with a staining quality rating from 1 (unacceptable) to 4 (excellent). RESULTS: The AI model successfully transformed the WLI of polyps with different morphologies and different types of lesions in the gastrointestinal tract into virtual IC-dyed images. The quality ratings of the real IC-dyed and AI images did not significantly differ concerning surface structure (AI vs IC: 3.08 vs 3.00), lesion border (3.04 vs 2.98), and overall contrast (3.14 vs 3.02) from 10 sets of images (10 AI images and 10 real IC-dyed images). Although the score depended significantly on the evaluator, the staining methods (AI or real IC) and evaluators had no significant interaction (P > 0.05) with each other. CONCLUSION: Our results demonstrated the feasibility of employing AI model's virtual IC staining, increasing the possibility of being employed in daily practice. This novel technology may facilitate gastrointestinal lesion identification in the future.

Novel screening model of obstructive sleep apnea for snorers with suspected NAFLD undergoing liver sonography.

BACKGROUND/AIMS: Given the increased incidence of obstructive sleep apnea (OSA) among patients with nonalcoholic fatty liver disease (NAFLD), noninvasive screening methods are urgently needed to screen for OSA risk in these patients when conducting an office-based assessment of hepatic steatosis. Therefore, we investigated the controlled attenuation parameter (CAP) and hepatic steatosis index (HSI) in patients with and without OSA and developed screening models to detect OSA. METHODS: We retrospectively reviewed the medical records of all adult snorers with suspected NAFLD undergoing liver sonography between June 2017 and June 2020. Records encompassed CAP and HSI data as well as data collected during in-hospital full-night polysomnography. The multivariate logistic regression models were constructed to explore the predictors of OSA risk. Furthermore, model validation was performed based on the medical records corresponding to the July 2020-June 2021 period. RESULTS: A total of 59 patients were included: 81.4% (48/59) were men, and the mean body mass index (BMI) was 26.4 kg/m2. Among the patients, 62.7% (37/59) and 74.6% (44/59) (detected by the HSI and CAP, respectively) had NAFLD, and 78% (46/59) were diagnosed with OSA on the basis of polysomnography. Three screening models based on multivariate analysis were established. The model combining male sex, a BMI of > 24.8, and an HSI of > 38.3 screened for OSA risk the most accurately, with an area under the receiver operating characteristic curve of 0.81 (sensitivity: 78%; specificity: 85%; and positive and negative predictive values: 95% and 52%, respectively) in the modeling cohort. An accuracy of 70.0% was achieved in the validation group. CONCLUSIONS: The combination screening models proposed herein provide a convenient, noninvasive, and rapid screening tool for OSA risk and can be employed while patients receive routine hepatic check-ups. These models can assist physicians in identifying at-risk OSA patients and thus facilitate earlier detection and timely treatment initiation.

Proteomic analysis of serum extracellular vesicles from biliary tract infection patients to identify novel biomarkers.

Biliary tract infection (BTI), a commonly occurring abdominal disease, despite being extensively studied for its initiation and underlying mechanisms, continues to pose a challenge in the quest for identifying specific diagnostic biomarkers. Extracellular vesicles (EVs), which emanate from diverse cell types, serve as minute biological entities that mirror unique physiological or pathological conditions. Despite their potential, there has been a relatively restricted exploration of EV-oriented methodologies for diagnosing BTI. To uncover potent protein biomarkers for BTI patients, we applied a label-free quantitative proteomic method known for its unbiased and high-throughput nature. Furthermore, 192 differentially expressed proteins surfaced within EVs isolated from individuals afflicted with BTI. Subsequent GO and KEGG analyses pinpointed Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) and Crumbs homolog 3 (CRB3) as noteworthy biomarkers. Validation via data analysis of plasma-derived EV samples confirmed their specificity to BTI. Our study leveraged an unbiased proteomic tool to unveil CEACAM1 and CRB3 as promising protein biomarkers in serum EVs, presenting potential avenues for the advancement of diagnostic systems for BTI detection.

The influence of early adoption of nonenhanced computed tomography on management of patients with pyogenic liver abscess.

Background and Aim: A pyogenic liver abscess (PLA) is an infectious disease with high in-hospital mortality. It has no specific symptoms and is difficult to be diagnosed early in the emergency department. Ultrasound is commonly used to detect PLA lesions of PLA, but its sensitivity can be affected by lesion size, location, and clinician experience. Therefore, early diagnosis and prompt treatment (especially abscess drainage) are crucial for better patient outcomes and should be prioritized by clinical physicians. Methods: We conducted a retrospective study to compare the effect of early and late (i.e., receiving CT scanning within 48 h and >48 h after admission) adoption of nonenhanced computed tomography (CT) scanning regarding the hospitalization days and interval between admission and drainage of patients with PLA. Results: This study included 76 hospitalized patients with PLA in the Department of Digestive Disease of Xiamen Chang Gung Hospital in China who underwent CT examinations from 2014 to 2021. We conducted CT scans on 56 patients within 48 h of admission and on 20 patients more than 48 h after admission. The early CT group had a significantly shorter hospitalization length compared with the late CT group (15.0 days vs. 20.5 days; P = 0.035). Besides, the median time to initiate drainage after admission was also significantly shorter in the early CT group than in the late CT group (1.0 days vs. 4.5 days; P < 0.001). Conclusion: Early CT scanning within 48 h of admission may aid in early PLA diagnosis and benefit disease recovery, as revealed by our findings.

Electroencephalogram pattern predicting neurological outcomes of children with seizures secondary to abusive head trauma.

BACKGROUND: The clinical presentations of abusive head trauma can abruptly worsen, so the occurrence of seizures and changes of EEG can be variable according to patients' conditions. Since the changes of EEG background waves reflect the cortical function of children, we aimed to find out whether the timing of EEG background, epileptiform discharges and seizure patterns were associated with the outcomes of patients with AHT. MATERIAL AND METHODS: Using seizure type and acute stage electroencephalographic (EEG) characteristics to assess adverse neurological outcomes in children with seizures secondary to abusive head trauma (AHT). Children who were hospitalized with AHT at a tertiary referral hospital from October 2000 to April 2010 were evaluated retrospectively. A total of 50 children below 6 years of age admitted due to AHT were included. KOSCHI outcome scale was used to evaluate the primary outcome and neurological impairment was used as secondary outcome after 6 months discharge. RESULTS: Children with apnea, cardiac arrest, reverse blood flow and skull fracture in clinic had a higher mortality rate even in the no-seizure group (3/5 [60%] vs. 3/45 [6.7%], odds ratio [OR] = 11; 95% CI = 2.3-52; p = 0.025). Seizure occurrence reduced mostly at the second day after admission in seizure groups; but children with persistent seizures for 1 week showed poor neurological outcomes. The occurrence of initial seizure was frequency associated with younger age; focal seizure, diffuse cortical dysfunction in acute-stage EEG, and low Glasgow Coma Scale (GCS) score were significantly related to poor outcomes after 6 months. Diffuse cortical dysfunction was also associated with motor, speech, and cognitive dysfunction. CONCLUSIONS: Diffuse cortical dysfunction in acute-stage EEG combined with low GCS score and focal seizure may related to poor outcomes and neurological dysfunctions in children with AHT.

Intestinal fatty acid-binding protein is a biomarker for diagnosis of biliary tract infection.

BACKGROUND AND AIM: Biliary tract infection (BTI) is an inflammatory disease and commonly associated with bacteremia. Delays in diagnosis or treatment of BTI cause high morbidity and mortality. However, an early diagnosis depends on appropriate clinical investigations. Appropriate biomarkers are urgently needed to improve the BTI diagnostic rate. We hypothesized that intestinal fatty acid-binding protein (I-FABP) might be a potential biomarker for BTI diagnosis. METHODS: We examined data from subjects aged ≥18 years diagnosed with BTI, including cholangitis and cholecystitis, whose blood samples were adequate for I-FABP and zonulin assessment. We also collected blood samples from healthy volunteers as the control group. We excluded subjects in both groups who received steroids, antibiotics, or probiotics within 1 month before hospital admission (BTI cohort) or participation in this research (controls). The main study endpoint was to compare the diagnostic ability of I-FABP to detect BTI in comparison with high-sensitivity C-reactive protein (hs-CRP) and zonulin. RESULTS: The study collected the data of 51 patients with BTI and 35 healthy subjects. The receiver operating characteristic (ROC) area under the curve (AUC) for I-FABP was 0.884 (95% confidence interval [CI]: 0.814-0.954), numerically higher than that for hs-CRP (0.880; 0.785-0.976) and zonulin (0.570; 0.444-0.697). We estimated that the optimal cutoff value of I-FABP was 2.1 ng/mL (sensitivity: 0.804; specificity: 0.829) for the diagnosis of BTI. CONCLUSIONS: In summary, this study suggests that I-FABP may be a potential alternative biomarker to hs-CRP for diagnosing BTI. Further research should verify the use of I-FABP as a marker for BTI diagnosis, but also for other inflammatory diseases.