RESUMEN
Ultrasound image guidance is a method often used to help provide care, and it relies on accurate perception of information, and particularly tissue recognition, to guide medical procedures. It is widely used in various scenarios that are often complex. Recent breakthroughs in large models, such as ChatGPT for natural language processing and Segment Anything Model (SAM) for image segmentation, have revolutionized interaction with information. These large models exhibit a revolutionized understanding of basic information, holding promise for medicine, including the potential for universal autonomous ultrasound image guidance. The current study evaluated the performance of SAM on commonly used ultrasound images and it discusses SAM's potential contribution to an intelligent image-guided framework, with a specific focus on autonomous and universal ultrasound image guidance. Results indicate that SAM performs well in ultrasound image segmentation and has the potential to enable universal intelligent ultrasound image guidance.
Asunto(s)
Algoritmos , UltrasonografíaRESUMEN
Background: Long-term treatment-free remission (TFR) represents a new goal for chronic myeloid leukemia (CML). Optimizing dose of tyrosine kinase inhibitors (TKIs) in the CML treatment maybe a new challenge to maintain effective and improving patients' quality of life. We hypothesized that administration of low-dose TKIs does not compromise major molecular response (MMR) in patients with CML who have a deep molecular response (DMR). Methods: We did an open-label, randomized trial at eight hospitals in China. Eligible CML-CP patients (aged 18-70 years) had shown continuous response to TKI more than 5 years and maintained MR4.5 (BCR-ABLIS ≤ 0.0032%) in recent 18 months. Patients were randomly assigned (1:1) to the TKI de-escalation group or the discontinuation group. Randomization was done with permuted blocks (block size four) and implemented through an interactive web-based randomization system. Recurrence was defined as the single sample with real time Quantitative PCR (RT-qPCR) measurement greater than 0.1% (MMR). The primary endpoint was 12-month MMR rate in patients who received de-escalation or discontinuation of TKIs. This study was registered at ClinicalTrials.gov (NCT04143087). Results: Around 125 patients were enrolled between October 23, 2019 and October 31, 2020, 62 patients received dose de-escalation of TKIs, while 63 patients in the discontinuation group. In the de-escalation group, molecular recurrence-free survival at 12 months was 88.32% (95% CI 79%-98%), whereas molecular recurrence-free survival in the discontinuation group at 12 months was 59.98% (95% CI 47-73). No progressions occurred at the data cut-off date. All 29 recurrence cases restart TKI treatment returned to MMR. Cytolytic NK cells as a proportion of lymphocyte cells were significantly increased from baseline after 6 months whether in the de-escalation or TKIs cessation group (P = 0.048, 0.001, respectively); compared with the relapsing patients, Tregs proportion was decreased (P = 0.003), and higher proportion of NK cells were found in non-relapsing patients whether in TKI de-escalation or discontinuation group (P = 0.011, 0.007, respectively). We also found that the de-escalation group showed better disease-specific HRQOL in regards to its impact on emotional functioning, fatigue, pain, and financial difficulties. Conclusion: With 88.32% MMR in 12-months follow-up after de-escalation TKIs' treatment, dose-halving could become a new treatment paradigm for CML patients who with DMR under continuing maintenance therapy with TKIs.
RESUMEN
Background: Acute myelogenous leukemia (AML) is a common and fatal disease in hematology with frequent relapses and a poor prognosis. Pyroptosis, a programmed cell death mediated by inflammasomes, has been shown to be associated with leukemia recently. However, the role of pyroptosis for diagnosis and prognosis in AML remained less understood. Methods: We downloaded three public datasets and constructed a signature of TCGA cohort using the least absolute shrinkage and selection operator (LASSO) Cox regression model to predict the overall survival of AML patients. Samples from the GEO database were treated as a validation cohort. Gone through LASSO-Cox regression analysis, an 8-PRG-related signature was developed. Used the median score from the signature, we classified patients in two subgroups. Subsequently, we employed univariate COX, multivariate Cox regression, ROC analysis and constructed a nomogram, Finally, differential analysis, GO and KEGG functional analysis, ESTIMATE algorithm and CIBERSORT algorithm were used to explore the difference between two groups. Results: The expression levels of 90.9% pyroptosis-related genes (PRGs) had significant difference compared AML with normal tissues. The results of univariate COX regression analysis demonstrated 10 differentially expressed genes (DEGs) were associated with patients' OS (p < 0.05). Then, we found OS of patients in the low-risk group was more likely to be lengthened compared with their high-risk counterparts (P < 0.05 both in the TCGA and GEO cohort). After controlling clinical factors, the risk score could still remain an independent predictive element (HR > 1, P < 0.001) of OS in multivariate Cox regression analysis. Furthermore, a nomogram with prognostic value for AML was thus established. Time-dependent ROC analysis proved the predictive power of the signature. Functional analysis suggested that DEGs were mainly concentrated in immune-related pathways, such as humoral immune response and T cell proliferation. TME scores and risk scores were strongly correlated and immune status differed between the risk subgroups. Conclusion: A novel PRG-related signature was established to forecast the prognosis in AML, and pyroptosis may be a potential therapeutic target for AML.
