Dual domain distribution disruption with semantics preservation: Unsupervised domain adaptation for medical image segmentation.

Recent unsupervised domain adaptation (UDA) methods in medical image segmentation commonly utilize Generative Adversarial Networks (GANs) for domain translation. However, the translated images often exhibit a distribution deviation from the ideal due to the inherent instability of GANs, leading to challenges such as visual inconsistency and incorrect style, consequently causing the segmentation model to fall into the fixed wrong pattern. To address this problem, we propose a novel UDA framework known as Dual Domain Distribution Disruption with Semantics Preservation (DDSP). Departing from the idea of generating images conforming to the target domain distribution in GAN-based UDA methods, we make the model domain-agnostic and focus on anatomical structural information by leveraging semantic information as constraints to guide the model to adapt to images with disrupted distributions in both source and target domains. Furthermore, we introduce the inter-channel similarity feature alignment based on the domain-invariant structural prior information, which facilitates the shared pixel-wise classifier to achieve robust performance on target domain features by aligning the source and target domain features across channels. Without any exaggeration, our method significantly outperforms existing state-of-the-art UDA methods on three public datasets (i.e., the heart dataset, the brain dataset, and the prostate dataset). The code is available at https://github.com/MIXAILAB/DDSPSeg.

Deactivation Kinetics of Polyethylenimine-based Adsorbents Used for the Capture of Low Concentration CO2.

CO2 emission is generally regarded as the major contributor to global climate change, and polyethylenimine (PEI)-based CO2 adsorbents are promising materials for the capture of low concentration CO2. This paper deals with the deactivation kinetics of PEI-based CO2 adsorbents used for the capture of low concentration CO2. EA and TG analyses demonstrated that thermal degradation and O2-induced deactivation of the adsorbents occurred simultaneously under air exposure conditions. It was found by N2 exposure experiments at the temperature of 50-80 °C that the thermal degradation of PEI-based adsorbents followed a first-order reaction model with an activation energy of 80.98 kJ/mol and a pre-exponential factor of 6.055 × 108 (h-1). The parallel reaction model was employed to distinguish the O2-induced deactivation from the thermal degradation of the adsorbents through air exposure experiments within 50-80 °C. The O2-induced deactivation exhibited a second-order reaction with an activation energy of 74.47 kJ/mol and a pre-exponential factor of 6.321 × 106 (%-1·h-1). The results of simulating the overall deactivation of the adsorbents by the parallel reaction kinetic model were well consistent with those of the experiments, proving that the parallel reaction model was feasible for the description of the deactivation of PEI-based adsorbents.

Prediction of Setschenow constants of N-heteroaromatics in NaCl solutions based on the partial charge on the heterocyclic nitrogen atom.

The solubilities of 19 different kinds of N-heteroaromatic compounds in aqueous solutions with different concentrations of NaCl were determined at 298.15 K with a UV-vis spectrophotometry and titration method, respectively. Setschenow constants, Ks, were employed to describe the solubility behavior, and it is found that the higher ring numbers of N-heteroaromatics gave rise to the lower values of Ks. Moreover, Ks showed a good linear relationship with the partial charge on the nitrogen atom (QN) for either QN > 0 or QN < 0 N-heteroaromatics. It further revealed that QN was well-matched in the prediction of salting-out effect for N-heteroaromatics compared to the conventional descriptors such as molar volume (VH) and the octanol-water partition coefficient (Kow). The heterocyclic N in N-heteroaromatics may interact with Na(+) ions in NaCl solution for QN < 0 and with Cl(-) for QN > 0.

Simultaneous Online Measurement of H2O and CO2 in the Humid CO2 Adsorption/Desorption Process.

A dew point meter (DP) and an infrared (IR) CO2 analyzer were assembled in a humid CO2 adsorption/desorption system in series for simultaneous online measurements of H2O and CO2, respectively. The humidifier, by using surface-flushing on a saturated brine solution was self-made for the generation of humid air flow. It was found that by this method it became relatively easy to obtain a low H2O content in air flow and that its fluctuation could be reduced compared to the bubbling method. Water calibration for the DP-IR detector is necessary to be conducted for minimizing the measurement error of H2O. It demonstrated that the relative error (RA) for simultaneous online measurements H2O and CO2 in the desorption process is lower than 0.1%. The high RA in the adsorption of H2O is attributed to H2O adsorption on the transfer pipe and amplification of the measurement error. The high accuracy of simultaneous online measurements of H2O and CO2 is promising for investigating their co-adsorption/desorption behaviors, especially for direct CO2 capture from ambient air.

Measurement and ANN prediction of pH-dependent solubility of nitrogen-heterocyclic compounds.

Based on the solubility of 25 nitrogen-heterocyclic compounds (NHCs) measured by saturation shake-flask method, artificial neural network (ANN) was employed to the study of the quantitative relationship between the structure and pH-dependent solubility of NHCs. With genetic algorithm-multivariate linear regression (GA-MLR) approach, five out of the 1497 molecular descriptors computed by Dragon software were selected to describe the molecular structures of NHCs. Using the five selected molecular descriptors as well as pH and the partial charge on the nitrogen atom of NHCs (QN) as inputs of ANN, a quantitative structure-property relationship (QSPR) model without using Henderson-Hasselbalch (HH) equation was successfully developed to predict the aqueous solubility of NHCs in different pH water solutions. The prediction model performed well on the 25 model NHCs with an absolute average relative deviation (AARD) of 5.9%, while HH approach gave an AARD of 36.9% for the same model NHCs. It was found that QN played a very important role in the description of NHCs and, with QN, ANN became a potential tool for the prediction of pH-dependent solubility of NHCs.

Karyotypic evolution and tumor progression in head and neck squamous cell carcinomas.

Cytogenetic analysis was performed on primary tumors, and paired recurrent or metastatic lesions, in 14 patients with head and neck squamous cell carcinomas (HNSCC), in order to identify chromosomal aberrations associated with tumor initiation and progression. Abnormal karyotypes were found in 12 of the 14 patients, with distinctive karyotypic similarities shown in all informative pairs. For individual patients, the degree of karyotypic complexity was similar for the primaries and paired recurrent or metastatic lesions. All 22 samples with clonal chromosomal aberrations displayed complex karyotypes with multiple numerical and unbalanced structural rearrangements, resulting in extensive genomic imbalances. The pathway of clonal evolution could be traced in a few patients, supporting the notion that some aberrations or imbalances, particularly partial or entire loss of 3p, i(8q), and homogeneously staining regions commonly mapping to 11q13, were early genetic events in the initiation of HNSCC.