A machine learning model integrating clinical-radiomics-deep learning features accurately predicts postoperative recurrence and metastasis of primary gastrointestinal stromal tumors.

Objectives: Post-surgical prediction of recurrence or metastasis for primary gastrointestinal stromal tumors (GISTs) remains challenging. We aim to develop individualized clinical follow-up strategies for primary GIST patients, such as shortening follow-up time or extending drug administration based on the clinical deep learning radiomics model (CDLRM).

Methods: The clinical information on primary GISTs was collected from two independent centers.

[Effects of organic inputs on soil nutrients and microbial metabolism in newly reclaimed farmlands].

It remains unclear how soil microbial metabolism responds to organic input and the driving factors during soil carbon fixation and fertilization in newly reclaimed farmlands. We conducted a field experiment to explore the effects of different organic inputs on soil nutrients, organic carbon fractions, extracellular enzyme activities, microbial metabolism, and microbial carbon utilization efficiency in a newly reclaimed farmland with a rice-wheat rotation in Jiande of Zhejiang in 2022. Five treatments were implemented with equivalent C return in addition to conventional chemical fertilizers (NPK): 1) NPK alone (control), 2) NPK + manure + maize straw (MS), 3) NPK + manure (M), 4) NPK + straw biochar-based manure (MBF), and 5) NPK + straw biochar (MB).

Construction of a multiscale feature fusion model for indoor scene recognition and semantic segmentation.

Fast and accurate extraction of multi-scale semantic information is crucial for indoor applications such as navigation and personalized services. This paper focuses on large-scale indoor scene categories and small-scale indoor element semantics, proposing an RGB-based dual-task model for scene recognition and semantic segmentation. The model employs a pyramid pooling module to create a shared feature layer for both tasks.