We examined the feasibility of explainable computer-aided detection of cardiomegaly in routine clinical practice using segmentation-based methods. Overall, 793 retrospectively acquired posterior–anterior (PA) chest X-ray images (CXRs) of 793 patients were used to train deep learning (DL) models for lung and heart segmentation. The training dataset included PA CXRs from two public datasets and in-house PA CXRs. Two fully automated segmentation-based methods using state-of-the-art DL models for lung and heart segmentation were developed. The diagnostic performance was assessed and the reliability of the automatic cardiothoracic ratio (CTR) calculation was determined using the mean absolute error and paired t-test. The effects of thoracic pathological conditions on performance were assessed using subgroup analysis. One thousand PA CXRs of 1000 patients (480 men, 520 women; mean age 63 ± 23 years) were included. The CTR values derived from the DL models and diagnostic performance exhibited excellent agreement with reference standards for the whole test dataset. Performance of segmentation-based methods differed based on thoracic conditions. When tested using CXRs with lesions obscuring heart borders, the performance was lower than that for other thoracic pathological findings. Thus, segmentation-based methods using DL could detect cardiomegaly; however, the feasibility of computer-aided detection of cardiomegaly without human intervention was limited.

The Systemic Inflammation Response Index (SIRI) has been used to predict the prognosis of various cancers. This study examined SIRI as a prognostic factor in the neoadjuvant setting and determined whether it changing after chemotherapy is related to patient prognosis. Patients who underwent pancreatic surgery following neoadjuvant chemotherapy for pancreatic cancer were retrospectively analyzed. To establish the cut-off values, SIRIpre-neoadjuvant, SIRIpost-neoadjuvant, and SIRIquotient (SIRIpost-neoadjuvant/SIRIpre-neoadjuvant) were calculated and significant SIRI values were statistically determined to examine their effects on survival rate. The study included 160 patients. Values of SIRIpost-neoadjuvant ≥ 0.8710 and SIRIquotient <0.9516 affected prognosis (hazard ratio [HR], 1.948; 95% confidence interval [CI], 1.210-3.135; ∗∗P = 0.006; HR, 1.548; 95% CI, 1.041-2.302; ∗∗P = 0.031). Disease-free survival differed significantly at values of SIRIpost-neoadjuvant < 0.8710 and SIRIpost-neoadjuvant ≥ 0.8710 (P = 0.0303). Overall survival differed significantly between SIRIquotient <0.9516 and SIRIquotient ≥0.9516 (P = 0.0368). SIRI can predict the survival of patients with pancreatic ductal adenocarcinoma after resection and neoadjuvant chemotherapy. Preoperative SIRI value was correlated with disease-free survival, while changes in SIRI values were correlated with overall survival.

Purpose: This study attempted to analyze the prognostic role of single nucleotide polymorphism array (SNP-A) –based karyotying in 133 patients with acute myeloid leukemia with normal karyotype (AML-NK), which presents with diverse clinical outcomes, thus requiring further stratification of patient subgroups according to their prognoses.Patients and Methods: A total of 133 patients with AML-NK confirmed by metaphase cytogenetics (MC) and fluorescent in situ hybridization analysis were included in this study. Analysis by Genome-Wide Human SNP 6.0 Array was performed by using DNAs derived from marrow samples at diagnosis.Results: Forty-three patients (32.3%) had at least one abnormal SNP lesion that was not detected by MC. One hundred thirteen abnormal SNP lesions included 55 losses, 23 gains, and 35 copy-neutral losses of heterozygosity. Multivariate analyses showed that detection of abnormal SNP lesions by SNP-A karyotyping results in an unfavorable prognostic value for overall survival (hazard ratio [HR], 2.69; 95% CI, 1.50 to 4.82; P = .001); other significant prognostic factors included secondary AML (HR, 5.55; 95% CI, 1.80 to 17.14; P = .003), presence of the FLT3 mutation (HR, 3.17; 95% CI, 1.71 to 5.87; P < .001), and age (HR, 1.03; 95% CI, 1.01 to 1.05; P = .020).Conclusion: Our data demonstrated that abnormal SNP lesions detected by SNP-A karyotyping might indicate an adverse prognosis in patients with AML-NK, thus requiring a more sophisticated treatment strategy for improvement of treatment outcomes.