4.4 kW Nd:YAG slab amplifier with a high repetition frequency and high beam quality.

Using the self-developed fused indium wetting technology and planar waveguide, the uniform heat dissipation of the slab crystal and uniform pumping of the pump light were achieved, respectively. Based on the master oscillator power amplification (MOPA) scheme, the power was then amplified when the seed light source passed through the Nd:YAG slab crystal three times. Additionally, the image transfer system that we added to the amplified optical path achieved high beam quality. Finally, we obtained a rectangular pulsed laser with an output average power of 4461 W, a repetition frequency of 20 kHz, a pulse width of 62 ns, an optical-to-optical conversion efficiency of 26.8%, and a beam quality of ß x=7.0 and ß y=7.7.

Deep learning system for malignancy risk prediction in cystic renal lesions: a multicenter study.

OBJECTIVES: To develop an interactive, non-invasive artificial intelligence (AI) system for malignancy risk prediction in cystic renal lesions (CRLs). METHODS: In this retrospective, multicenter diagnostic study, we evaluated 715 patients. An interactive geodesic-based 3D segmentation model was created for CRLs segmentation. A CRLs classification model was developed using spatial encoder temporal decoder (SETD) architecture. The classification model combines a 3D-ResNet50 network for extracting spatial features and a gated recurrent unit (GRU) network for decoding temporal features from multi-phase CT images. We assessed the segmentation model using sensitivity (SEN), specificity (SPE), intersection over union (IOU), and dice similarity (Dice) metrics. The classification model's performance was evaluated using the area under the receiver operator characteristic curve (AUC), accuracy score (ACC), and decision curve analysis (DCA). RESULTS: From 2012 to 2023, we included 477 CRLs (median age, 57 [IQR: 48-65]; 173 men) in the training cohort, 226 CRLs (median age, 60 [IQR: 52-69]; 77 men) in the validation cohort, and 239 CRLs (median age, 59 [IQR: 53-69]; 95 men) in the testing cohort (external validation cohort 1, cohort 2, and cohort 3). The segmentation model and SETD classifier exhibited excellent performance in both validation (AUC = 0.973, ACC = 0.916, Dice = 0.847, IOU = 0.743, SEN = 0.840, SPE = 1.000) and testing datasets (AUC = 0.998, ACC = 0.988, Dice = 0.861, IOU = 0.762, SEN = 0.876, SPE = 1.000). CONCLUSION: The AI system demonstrated excellent benign-malignant discriminatory ability across both validation and testing datasets and illustrated improved clinical decision-making utility. CRITICAL RELEVANCE STATEMENT: In this era when incidental CRLs are prevalent, this interactive, non-invasive AI system will facilitate accurate diagnosis of CRLs, reducing excessive follow-up and overtreatment. KEY POINTS: The rising prevalence of CRLs necessitates better malignancy prediction strategies. The AI system demonstrated excellent diagnostic performance in identifying malignant CRL. The AI system illustrated improved clinical decision-making utility.

Dual-Channel Prototype Network for Few-Shot Pathology Image Classification.

In the field of pathology, the scarcity of certain diseases and the difficulty of annotating images hinder the development of large, high-quality datasets, which in turn affects the advancement of deep learning-assisted diagnostics. Few-shot learning has demonstrated unique advantages in modeling tasks with limited data, yet explorations of this method in the field of pathology remain in the early stages. To address this issue, we present a dual-channel prototype network (DCPN), a novel few-shot learning approach for efficiently classifying pathology images with limited data. The DCPN leverages self-supervised learning to extend the pyramid vision transformer (PVT) to few-shot classification tasks and combines it with a convolutional neural network to construct a dual-channel network for extracting multi-scale, high-precision pathological features, thereby substantially enhancing the generalizability of prototype representations. Additionally, we design a soft voting classifier based on multi-scale features to further augment the discriminative power of the model in complex pathology image classification tasks. We constructed three few-shot classification tasks with varying degrees of domain shift using three publicly available pathological datasets-CRCTP, NCTCRC, and LC25000-to emulate real-world clinical scenarios. The results demonstrated that the DCPN outperformed the prototypical network across all metrics, achieving the highest accuracies in same-domain tasks-70.86% for 1-shot, 82.57% for 5-shot, and 85.2% for 10-shot setups-corresponding to improvements of 5.51%, 5.72%, and 6.81%, respectively, over the prototypical network. Notably, in the same-domain 10-shot setting, the accuracy of the DCPN (85.2%) surpassed that of the PVT-based supervised learning model (85.15%), confirming its potential to diagnose rare diseases within few-shot learning frameworks.

Bioactive neolignan, iridoid and flavonoid glycosides from the leaves of Vaccinium bracteatum.

Two neolignan glycosides including a new one (1), along with seven iridoid glycosides (3 - 9) and nine flavonoid glycosides (10 - 18), were isolated from the leaves of Vaccinium bracteatum. Their structures were established mainly on the basis of 1D/2D NMR and ESIMS analyses, as well as comparison to known compounds in the literature. The structure of 1 with absolute stereochemistry was also confirmed by chemical degradation and ECD calculation. Selective compounds showed antiradical activity against ABTS and/or DPPH. Moreover, several isolates also suppressed the production of ROS in RAW264.7 cells and exerted neuroprotective effect toward PC12 cells.

Bioactive monoterpenoids and acetophenones from the aerial parts of Eupatorium fortunei.

Thirty-nine thymol and acetophenone derivatives, including eight pairs of enantiomers, were isolated from the aerial parts of Eupatorium fortunei. Their structures were assigned by detailed analyses of spectroscopic data and NMR calculations based on density functional theory, with 18 ones (1a/1b-14) being previously undescribed compounds. While the absolute configurations of 1a/1b, 2a/2b, 4, 6a/6b, 7, 11a/11b and 15a/15b-18a/18b were established by calculations of electronic circular dichroism data, that of 14 was determined by modified Mosher's method. Compounds 1a/1b and 2a/2b represent a previously unreported type of monoterpenoid dimers via an amide linkage, and compound 3 is a monoterpene-phenylpropanoid hybrid connected through an ester bond. Among the known molecules, the formerly mis-assigned structures of 15a/15b and 22 were revised, and pure natural enantiomers of 16a/16b-18a/18b were reported for the first time. Selective compounds showed antiradical and NO production inhibitory activities in the preliminary biological screening. Compound 31 was further demonstrated to alleviate oxidative stress by activating Nrf2 signaling pathway.

Monoterpene indole alkaloids from Gelsemium elegans.

Two new monoterpene indole alkaloids, Eleganine A (1) and Eleganine B (2), along with 11 known compounds (3-13) were isolated from the stems and leaves of Gelsemium elegans. Compound 1 is a gelsenicine-related monoterpenoid indole alkaloid possessing an iridoid unit. Their structures and absolute configurations of 1-2 were established by UV, IR, HR-ESI-MS, NMR spectroscopy, and electronic circular dichroism data analyses. All isolated compounds were evaluated for their anti-inflammatory and inhibiting glucose-induced mesanginal cell proliferation activities. None of them showed activity with IC50 far beyond 50 µM.

Chemical constituents from Eupatorium fortunei and their anti-inflammatory evaluation by in silico and experimental approaches.

The well-known aromatic and medicinal plant Eupatorium fortunei Turcz. is widely cultivated in China, and previous studies on its bioactive constituents mainly focus on the essential oil ingredients especially thymol derivatives. However, reports on other type of constituents and the potential application are lacking. In the present project, an intensive chemical fractionation on the aerial part extract of E. fortunei led to the isolation and identification of a series of fatty acid derivatives (lipids, 1a/1b-19) including seven pairs of previously undescribed enantiomers (1a/1b-7a/7b), as well as a lignan (brachangobinan A (BBA), 20) and two monoterpenes (8S/8R-9-hydroxythymol, 21a/21b). A preliminary biological evaluation of these compounds in a NO production inhibitory assay model demonstrated compound BBA as the most active one. Network pharmacology analysis was used to predict and explore the possible anti-inflammatory targets and mechanism of BBA, which revealed some potential inflammation-related proteins and signaling pathways. Further experimental investigations validated that the anti-inflammatory effect of BBA could be achieved by suppressing pro-inflammatory factors and blocking the activation of NF-κB signaling pathway. Taken together, our work shows that E. fortunei can serve as a potential resource of lipids and anti-inflammatory agents.

Pyramid-based self-supervised learning for histopathological image classification.

Large-scale labeled datasets are crucial for the success of supervised learning in medical imaging. However, annotating histopathological images is a time-consuming and labor-intensive task that requires highly trained professionals. To address this challenge, self-supervised learning (SSL) can be utilized to pre-train models on large amounts of unsupervised data and transfer the learned representations to various downstream tasks. In this study, we propose a self-supervised Pyramid-based Local Wavelet Transformer (PLWT) model for effectively extracting rich image representations. The PLWT model extracts both local and global features to pre-train a large number of unlabeled histopathology images in a self-supervised manner. Wavelet is used to replace average pooling in the downsampling of the multi-head attention, achieving a significant reduction in information loss during the transmission of image features. Additionally, we introduce a Local Squeeze-and-Excitation (Local SE) module in the feedforward network in combination with the inverse residual to capture local image information. We evaluate PLWT's performance on three histopathological images and demonstrate the impact of pre-training. Our experiment results indicate that PLWT with self-supervised learning performs highly competitive when compared with other SSL methods, and the transferability of visual representations generated by SSL on domain-relevant histopathological images exceeds that of the supervised baseline trained on ImageNet.

Monoterpene indole alkaloids from the roots of Alstonia rupestris and their anti-inflammatory activity.

Four new monoterpene indole alkaloids (1-4) together with twelve known alkaloids (5-16) were isolated from the roots of Alstonia rupestris. Compound 1 was the first example of C2-symmetric heteroyohimbine-type indole alkaloid homodimer obtained from natural plant resource. Their structures were elucidated on the basis of spectroscopic data. The absolute configuration of 1 was determined by comparison of its calculated and experimental electronic circular dichroism (ECD) spectra. All compounds were evaluated for their anti-inflammatory activities by measuring their NO inhibitory effects in LPS-stimulated RAW 264.7 cells. Compound 2 showed strong NO inhibition with IC50 value of 4.2 ± 1.3 µM. Moreover, compound 2 could decrease the expressions of cyclooxygenase-2 (COX-2) and transforming growth factor beta-1 (TGF-ß1).

Learning how to detect: A deep reinforcement learning method for whole-slide melanoma histopathology images.

Cutaneous melanoma represents one of the most life-threatening malignancies. Histopathological image analysis serves as a vital tool for early melanoma detection. Deep neural network (DNN) models are frequently employed to aid pathologists in enhancing the efficiency and accuracy of diagnoses. However, due to the paucity of well-annotated, high-resolution, whole-slide histopathology image (WSI) datasets, WSIs are typically fragmented into numerous patches during the model training and testing stages. This process disregards the inherent interconnectedness among patches, potentially impeding the models' performance. Additionally, the presence of excess, non-contributing patches extends processing times and introduces substantial computational burdens. To mitigate these issues, we draw inspiration from the clinical decision-making processes of dermatopathologists to propose an innovative, weakly supervised deep reinforcement learning framework, titled Fast medical decision-making in melanoma histopathology images (FastMDP-RL). This framework expedites model inference by reducing the number of irrelevant patches identified within WSIs. FastMDP-RL integrates two DNN-based agents: the search agent (SeAgent) and the decision agent (DeAgent). The SeAgent initiates actions, steered by the image features observed in the current viewing field at various magnifications. Simultaneously, the DeAgent provides labeling probabilities for each patch. We utilize multi-instance learning (MIL) to construct a teacher-guided model (MILTG), serving a dual purpose: rewarding the SeAgent and guiding the DeAgent. Our evaluations were conducted using two melanoma datasets: the publicly accessible TCIA-CM dataset and the proprietary MELSC dataset. Our experimental findings affirm FastMDP-RL's ability to expedite inference and accurately predict WSIs, even in the absence of pixel-level annotations. Moreover, our research investigates the WSI-based interactive environment, encompassing the design of agents, state and reward functions, and feature extractors suitable for melanoma tissue images. This investigation offers valuable insights and references for researchers engaged in related studies. The code is available at: https://github.com/titizheng/FastMDP-RL.

Cu-doped SnO2/rGO nanocomposites for ultrasensitive H2S detection at low temperature.

Hydrogen sulfide (H2S) detection remains a significant concern and the sensitivity, selectivity, and detection limit must be balanced at low temperatures. Herein, we utilized a facile solvothermal method to prepare Cu-doped SnO2/rGO nanocomposites that have emerged as promising candidate materials for H2S sensors. Characterization of the Cu-SnO2/rGO was carried out to determine its surface morphology, chemical composition, and crystal defects. The optimal sensor response for 10 ppm H2S was ~1415.7 at 120 °C, which was over 320 times higher than that seen for pristine SnO2 CQDs (Ra/Rg = 4.4) at 280 °C. Moreover, the sensor material exhibited excellent selectivity, a superior linear working range (R2 = 0.991, 1-150 ppm), a fast response time (31 s to 2 ppm), and ppb-level H2S detection (Ra/Rg = 1.26 to 50 ppb) at 120 °C. In addition, the sensor maintained a high performance even at extremely high humidity (90%) and showed outstanding long-term stability. These superb H2S sensing properties were attributed to catalytic sensitization by the Cu dopant and a synergistic effect of the Cu-SnO2 and rGO, which offered abundant active sites for O2 and H2S absorption and accelerated the transfer of electrons/holes.

Anti-inflammatory labdane diterpenoids from the aerial parts of Leonurus japonicus.

Twenty-two labdane-type diterpenoids, including ten pairs of 15-epimers and a pair of 13,15-epimers, were obtained from the aerial parts of a well-known medicinal plant Leonurus japonicus Houtt. While these epimers were separated by chiral HPLC, their structures were established mainly via spectroscopic methods especially NMR, X-ray crystallography and ECD techniques. Among them, seventeen compounds, encompassing three pairs of solvolysis artefacts likely due to the use of ethanol as extracting solvent, were reported for the first time in the current work. Our preliminary anti-inflammatory screening demonstrated that seven diterpenoids displayed noteworthy inhibitory effect on the NO production in LPS-induced RAW264.7 cells. In addition, the release of pro-inflammatory factors TNF-α, IL-1ß and IL-6, as well as the expression of iNOS and COX-2 proteins, was also suppressed by the unreported 15,16-epoxy-6ß-hydroxy-15α-methoxy-7,16-dioxolabd-8,13-diene. Further investigation into the preliminary anti-inflammatory mechanism of this compound indicated that it could block the activation of NF-κB signaling pathway.

Pd/Pt-Bimetallic-Nanoparticle-Doped In2O3 Hollow Microspheres for Rapid and Sensitive H2S Sensing at Low Temperature.

H2S is a poisonous gas that is widespread in nature and human activities. Its rapid and sensitive detection is essential to prevent it from damaging health. Herein, we report Pd- and Pt-bimetallic-nanoparticle-doped In2O3 hollow microspheres that are synthesized using solvothermal and in situ reduction methods for H2S detection. The structure of as-synthesized 1 at% Pd/Pt-In2O3 comprises porous hollow microspheres assembled from In2O3 nanosheets with Pd and Pt bimetallic nanoparticles loaded on its surface. The response of 1 at% Pd/Pt-In2O3 to 5 ppm H2S is 140 (70 times that of pure In2O3), and the response time is 3 s at a low temperature of 50 °C. In addition, it can detect trace H2S (as low as 50 ppb) and has superior selectivity and an excellent anti-interference ability. These outstanding gas-sensing performances of 1 at% Pd/Pt-In2O3 are attributed to the chemical sensitization of Pt, the electronic sensitization of Pd, and the synergistic effect between them. This work supplements the research of In2O3-based H2S sensors and proves that Pd- and Pt-bimetallic-doped In2O3 can be applied in the detection of H2S.

Deep learning and radiomic feature-based blending ensemble classifier for malignancy risk prediction in cystic renal lesions.

BACKGROUND: The rising prevalence of cystic renal lesions (CRLs) detected by computed tomography necessitates better identification of the malignant cystic renal neoplasms since a significant majority of CRLs are benign renal cysts. Using arterial phase CT scans combined with pathology diagnosis results, a fusion feature-based blending ensemble machine learning model was created to identify malignant renal neoplasms from cystic renal lesions (CRLs). Histopathology results were adopted as diagnosis standard. Pretrained 3D-ResNet50 network was selected for non-handcrafted features extraction and pyradiomics toolbox was selected for handcrafted features extraction. Tenfold cross validated least absolute shrinkage and selection operator regression methods were selected to identify the most discriminative candidate features in the development cohort. Feature's reproducibility was evaluated by intra-class correlation coefficients and inter-class correlation coefficients. Pearson correlation coefficients for normal distribution and Spearman's rank correlation coefficients for non-normal distribution were utilized to remove redundant features. After that, a blending ensemble machine learning model were developed in training cohort. Area under the receiver operator characteristic curve (AUC), accuracy score (ACC), and decision curve analysis (DCA) were employed to evaluate the performance of the final model in testing cohort. RESULTS: The fusion feature-based machine learning algorithm demonstrated excellent diagnostic performance in external validation dataset (AUC = 0.934, ACC = 0.905). Net benefits presented by DCA are higher than Bosniak-2019 version classification for stratifying patients with CRL to the appropriate surgery procedure. CONCLUSIONS: Fusion feature-based classifier accurately distinguished malignant and benign CRLs which outperformed the Bosniak-2019 version classification and illustrated improved clinical decision-making utility.

Methyl 2-naphthoates with anti-inflammatory activity from Morinda officinalis.

Chemical fractionation of the EtOH extract of the roots of a traditional Chinese herb, Morinda officinalis, afforded an array of methyl 2-naphthoate derivatives (1-9) including four pairs of enantiomers (1-4), two pimarane diterpenes and two ursane triterpenoids. Among them, eight compounds (1a/1b-3a/3b, 11 and 13) were reported in the current work for the first time. The structures of the new compounds, including their absolute configurations, were defined by spectroscopic analyses in combination with quantum chemical electronic circular dichroism (ECD) and gauge-independent atomic orbital (GIAO) NMR calculations. All the isolates were evaluated for their inhibitory effect on nitric oxide (NO) production induced by lipopolysaccharide (LPS) in murine RAW264.7 macrophage cells, and the enantiomers 1a and 3b exhibited moderate activity with IC50 values of 41.9 and 26.2 µM. Meanwhile, compound 3b also dose-dependently inhibited the secretion of two pro-inflammatory cytokines TNF-α and IL-6 in the same cell model.

Stratification of malignant renal neoplasms from cystic renal lesions using deep learning and radiomics features based on a stacking ensemble CT machine learning algorithm.

Using nephrographic phase CT images combined with pathology diagnosis, we aim to develop and validate a fusion feature-based stacking ensemble machine learning model to distinguish malignant renal neoplasms from cystic renal lesions (CRLs). This retrospective research includes 166 individuals with CRLs for model training and 47 individuals with CRLs in another institution for model testing. Histopathology results are adopted as diagnosis criterion. Nephrographic phase CT scans are selected to build the fusion feature-based machine learning algorithms. The pretrained 3D-ResNet50 CNN model and radiomics methods are selected to extract deep features and radiomics features, respectively. Fivefold cross-validated least absolute shrinkage and selection operator (LASSO) regression methods are adopted to identify the most discriminative candidate features in the development cohort. Intraclass correlation coefficients and interclass correlation coefficients are employed to evaluate feature's reproducibility. Pearson correlation coefficients for normal distribution features and Spearman's rank correlation coefficients for non-normal distribution features are used to eliminate redundant features. After that, stacking ensemble machine learning models are developed in the training cohort. The area under the receiver operator characteristic curve (ROC), calibration curve, and decision curve analysis (DCA) are adopted in the testing cohort to evaluate the performance of each model. The stacking ensemble machine learning algorithm reached excellent diagnostic performance in the testing dataset. The calibration plot shows good stability when using the stacking ensemble model. Net benefits presented by DCA are higher than the Bosniak 2019 version classification when employing any machine learning algorithm. The fusion feature-based machine learning algorithm accurately distinguishes malignant renal neoplasms from CRLs, which outperformed the Bosniak 2019 version classification, and proves to be more applicable for clinical decision-making.

Facial muscle movements in patients with Parkinson's disease undergoing phonation tests.

Purpose: Parkinson's disease (PD) is a serious neurodegenerative disease affecting the elderly. In general, the locomotion deficit, which seriously affects the daily life of patients with PD, usually occurs at a later stage. The mask face symptom meanwhile progressively worsens. However, facial muscle disorders and changes involved in the freezing mask are unclear. Method: In this study, we recruited 35 patients with PD and 26 age- and sex-balanced controls to undergo phonation tests, while the built-in camera on the laptop recorded their facial expressions during the whole pronunciation process. Furthermore, FaceReader (version 7.0; Noldus Information Technology, Wageningen, Netherlands) was used to analyze changes in PD facial landmark movement and region movement. Results: The two-tailed Student's t-test showed that the changes in facial landmark movement among 49 landmarks were significantly lower in patients with PD than in the control group (P < 0.05). The data on facial region movement revealed that the eyes and upper lip of patients with PD differed significantly from those in the control group. Conclusion: Patients with PD had defects in facial landmark movement and regional movement when producing a single syllable, double syllable, and multiple syllables, which may be related to reduced facial expressions in patients with PD.

POLIMI-ITW-S: A large-scale dataset for human activity recognition in the wild.

Human activity recognition is attracting increasing research attention. Many activity recognition datasets have been created to support the development and evaluation of new algorithms. Given the lack of datasets collected in real environments (In The Wild) to support human activity recognition in public spaces, we introduce a large-scale video dataset for activity recognition In The Wild: POLIMI-ITW-S. The fully labeled dataset consists of 22,161 RGB video clips (about 46 h) including 37 activity classes performed by 50 K+ subjects in real shopping malls. We evaluated the state-of-the-art models on this dataset and get relatively low accuracy. We release the dataset including the annotations composed by person tracking bounding boxes, 2-D skeleton, and activity labels for research use at: https://airlab.deib.polimi.it/polimi-itw-s-a-shopping-mall-dataset-in-the-wild.

Automatic CD30 scoring method for whole slide images of primary cutaneous CD30+ lymphoproliferative diseases.

AIMS: Deep-learning methods for scoring biomarkers are an active research topic. However, the superior performance of many studies relies on large datasets collected from clinical samples. In addition, there are fewer studies on immunohistochemical marker assessment for dermatological diseases. Accordingly, we developed a method for scoring CD30 based on convolutional neural networks for a few primary cutaneous CD30+ lymphoproliferative disorders and used this method to evaluate other biomarkers. METHODS: A multipatch spatial attention mechanism and conditional random field algorithm were used to fully fuse tumour tissue characteristics on immunohistochemical slides and alleviate the few sample feature deficits. We trained and tested 28 CD30+ immunohistochemical whole slide images (WSIs), evaluated them with a performance index, and compared them with the diagnoses of senior dermatologists. Finally, the model's performance was further demonstrated on the publicly available Yale HER2 cohort. RESULTS: Compared with the diagnoses by senior dermatologists, this method can better locate the tumour area and reduce the misdiagnosis rate. The prediction of CD3 and Ki-67 validated the model's ability to identify other biomarkers. CONCLUSIONS: In this study, using a few immunohistochemical WSIs, our model can accurately identify CD30, CD3 and Ki-67 markers. In addition, the model could be applied to additional tumour identification tasks to aid pathologists in diagnosis and benefit clinical evaluation.

Ultralow detection limit and ultrafast response/recovery of the H2 gas sensor based on Pd-doped rGO/ZnO-SnO2 from hydrothermal synthesis.

Hydrogen (H2) sensors are of great significance in hydrogen energy development and hydrogen safety monitoring. However, achieving fast and effective detection of low concentrations of hydrogen is a key problem to be solved in hydrogen sensing. In this work, we combined the excellent gas sensing properties of tin(IV) oxide (SnO2) and zinc oxide (ZnO) with the outstanding electrical properties of reduced graphene oxide (rGO) and prepared palladium (Pd)-doped rGO/ZnO-SnO2 nanocomposites by a hydrothermal method. The crystal structure, structural morphology, and elemental composition of the material were characterized by FE-SEM, TEM, XRD, XPS, Raman spectroscopy, and N2 adsorption-desorption. The results showed that the Pd-doped ZnO-SnO2 composites were successfully synthesized and uniformly coated on the surface of the rGO. The hydrogen gas sensing performance of the sensor prepared in this work was investigated, and the results showed that, compared with the pure Pd-doped ZnO-SnO2 sensor, the Pd-doped rGO/ZnO-SnO2 sensor modified with 3 wt% rGO had better hydrogen (H2)-sensing response of 9.4-100 ppm H2 at 380 °C. In addition, this sensor had extremely low time parameters (the response time and recovery time for 100 ppm H2 at 380 °C were 4 s and 8 s, respectively) and an extremely low detection limit (50 ppb). Moreover, the sensor exhibited outstanding repeatability and restoration. According to the analysis of the sensing mechanism of this nanocomposite, the enhanced sensing performance of the Pd-doped rGO/ZnO-SnO2 sensor is mainly due to the heterostructure of rGO, ZnO, and SnO2, the excellent electrical and physical properties of rGO and the synergy between rGO and Pd.