Research on duck egg recognition algorithm based on improved YOLOv4

1. The following study addressed the problem of small duck eggs as challenging to detect and identify for pick up in complex free-range duck farm environments. It introduces improvements to the YOLOv4 convolutional neural network target detection algorithm, based on the working conditions of egg-picking robots.2. Specifically, one scale of anchor boxes was removed from the prediction network, and a duck egg labelling dataset was established to make the improved algorithm YOLOv4-ours better match the working state of egg-picking robots and enhance detection performance.3. Through multiple comparative experiments, the YOLOv4-ours object detection algorithm exhibited superior overall performance, achieving a precision of 98.85%, recall of 96.67%, and an average precision of 98.60% and F1 score increased to 97%. Compared to the original YOLOv4 model, these improvements represented increases of 1.89%, 3.41%, 1.32%, and 1.04%, respectively. Furthermore, detection time was reduced from 0.26 seconds per image to 0.20 seconds.4. The enhanced model accurately detected duck eggs in free-range duck housing, effectively meeting the real-time egg identification and picking requirements.

Decoding the biology and clinical implication of neutrophils in intracranial aneurysm

OBJECTIVE

Abundant neutrophils have been identified in both ruptured and unruptured intracranial aneurysm (IA) domes, with their function and clinical implication being poorly characterized.

MATERIALS AND METHODS

We employed single-cell RNA sequencing (scRNA-Seq) datasets of both human and murine model, and external bulk mRNA sequencing datasets to thoroughly explore the features and functional heterogeneous of neutrophils infiltrating the IA dome.

RESULTS

We found that both unruptured and ruptured IA dome contain a substantial population of neutrophils, characterized by FCGR3B, G0S2, CSF3R, and CXCR2. These cells exhibited heterogeneity in terms of function and differentiation. Despite similar transcriptional activation, neutrophils in IA dome expressed a repertoire of gene programs that mimicked transcriptomic alterations observed from bone marrow to peripheral blood, showing self-similarity. In addition, the recruitment of neutrophils in unruptured IA was primarily mediated by monocytes/macrophages, and once ruptured, both neutrophils, and a specific subset of inflammatory smooth muscle cells (SMCs) were involved in the process. The receiver operator characteristic curve (ROC) analysis indicated that distinct neutrophil subclusters were associated with IA formation and rupture, respectively. By reviewing current studies, we found that neutrophils play a detrimental role to IA wall integrity through secreting specific ligands, ferroptosis driven by ALOX5AP and PTGS2, and the formation of neutrophil extracellular traps (NETs) mediated by PADI4.

INTERPRETATION

This study delineated the biology and potential clinical implications of neutrophils in IA dome and provided a reliable basis for future researches.

Management of cavernous sinus meningiomas: Clinical features, treatment strategies, and long-term outcomes

OBJECTIVES

The purpose of this research was to summarize the clinical and prognostic features of cavernous sinus meningiomas (CSM), evaluate the treatment strategies and long-term prognosis of CSM, and improve the management of CSM and the treatment effect for patients.

METHODS

We retrospectively studied the data of 54 patients who received initial surgical resection and 45 patients who received initial gamma knife radiosurgery (GKRS) for CSM at West China Hospital of Sichuan University from 2009 to 2021. Progression-free survival (PFS), Karnofsky Performance Scale (KPS) scores and neurological function recovery were adopted to assess a comprehensive management strategy for CSM.

RESULTS

Gross total resection (GTR) was performed in 51.9 % of cases with 3.7 % surgical mortality. The average follow-up time was 48.7 months, with a progression rate of 29.3 %. The overall improvement rate for cranial nerve function deficits was 50.0 %. By survival analysis, the extent of resection and the histological grade were significantly related to the prognosis. The role of postoperative GKRS is uncertain. For patients who received initial GKRS, the progression rate was 17.8 %, and the overall improvement rate for cranial nerve function deficits was 61.1 %. Primary treatment with GKRS showed better long-term tumor control in patients with CSM (P = 0.046).

CONCLUSIONS

Maximum safe resection of CSM can improve the neurological function and quality of life of patients, but aggressive resection will cause high perioperative mortality and complication rates. For CSM patients who are suitable for initial gamma knife treatment, choosing GKRS can achieve better long-term tumor control and neurological outcomes.

The Performance of Different Machine Learning Algorithm and Regression Models in Predicting High-Grade Intracranial Meningioma

Meningioma is the most common primary tumor of the central nervous system (CNS). Individualized treatment strategies should be formulated for the patients according to the WHO (World Health Organization) grade. Our aim was to investigate the effectiveness of various machine learning and traditional statistical models in predicting the WHO grade of preoperative patients with meningioma. Patients diagnosed with meningioma after surgery in West China Hospital and Shangjin Hospital of Sichuan University from 2009 to 2016 were included in the study cohort. As the training cohort (n = 1975), independent risk factors associated with high-grade meningioma were used to establish the Nomogram model. which was validated in a subsequent cohort (n = 1048) from 2017 to 2019 in our hospital. Logistic regression (LR), XGboost, Adaboost, Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest (RF) models were determined using F1 score, recall, accuracy, the area under the curve (ROC), calibration plot and decision curve analysis (DCA) were used to evaluate the different models. Logistic regression showed better predictive performance and interpretability than machine learning. Gender, recurrence history, T1 signal intensity, enhanced signal degree, peritumoral edema, tumor diameter, cystic, location, and NLR index were identified as independent risk factors and added to the nomogram. The AUC (Area Under Curve) value of RF was 0.812 in the training set, 0.807 in the internal validation set, and 0.842 in the external validation set. The calibration curve and DCA (Decision Curve Analysis) indicated that it had better prediction efficiency of LR than others. The Nomogram preoperative prediction model of meningioma of WHO II and III grades showed effective prediction ability. While machine learning exhibits strong fitting ability, it performs poorly in the validation set.

Clinical features, radiological findings, and prognostic factors for primary intracranial chordoid meningioma

OBJECTIVES

Chordoid meningioma (CM) is an infrequent histologic subtype of meningiomas. Owing to its low occurrence, this subtype has been rarely described. Our subject was to explore the clinical features, radiological characteristics, and prognostic factors of primary intracranial chordoid meningioma.

METHODS

We reviewed the medical records and collected follow-up information of 34 cases who had been surgically treated and histologically diagnosed with CM at the Department of Neurosurgery, West-China Hospital of Sichuan University, from January 2009 to December 2021.

RESULTS

Among all 7,950 meningioma cases, the proportion of primary intracranial CM was 0.43% (34/7,950). The median diagnosis age was 47 (ranging from 12 to 74) and the gender ratio (male to female) was 2.1:1. For radiological features, heterogeneous enhancement, skull base, and ventricular localization, cystic degeneration and dural tail sign were common in CM cases. In treatment, gross total resection (GTR) was achieved in 22/34 cases (64.7%) and subtotal resection (STR) was achieved in 12/34 cases (35.3%). Further, 11/34 patients (32.4%) had received postoperative adjuvant radiotherapy (RT). The follow-up duration ranged from 4 to 157 months after operation. The progression rate was 20.7% (6/29) and the median of PFS was 38 months. By survival analysis, accepting adjuvant radiotherapy and achieving GTR were correlated with longer progression-free survival for prognosis.

CONCLUSION

CM is a rare subtype of meningiomas. In our series, it mainly involved adults and did not show a predilection for women compared with meningiomas in general. For a better prognosis, gross total resection and postoperative adjuvant radiotherapy are recommended. Nevertheless, due to the restriction of the series sample, patients lost for follow-up and inherent biases of a retrospective study, more cases and a shorter follow-up duration are needed for better management of chordoid meningioma.

Management of Skull-Base Meningiomas With Extracranial Extensions: Clinical Features, Radiological Findings, Surgical Strategies, and Long-Term Outcomes

Objectives:

The aim of this research was to summarize the clinical and prognostic features of the skull-base meningiomas with extracranial extensions, and enhance the management of skull-base communicative meningiomas.

Methods

We retrospectively studied the medical records and analyzed the follow-up information of 53 patients who have done surgery for skull-base meningiomas with extracranial extensions in West China Hospital of Sichuan University from 2009 to 2020.

Results

The incidence of skull-base meningiomas with extracranial extensions was 0.74%. The average diagnosis age was 45.9 years, with a 1:3.1 men to women ratio. WHO grade I was seen in 84.9% of patients, and higher grades were found in 15.1%. Heterogeneous enhancement, high bone invasion rate, high incidence of peritumoral edema, and high dural tail sign rate were typical imaging features. Routine craniotomy and endoscopic endonasal approach were adopted, and gross total resection was performed in 62.3% of cases with 20.8% postoperative complication rates. The average follow-up time was 61.5 months, with a recurrence rate of 34.9%. By survival analysis, the extent of resection (p = 0.009) and the histological grade (p = 0.007) were significantly related to the prognosis. Adjuvant radiotherapy proved beneficial in patients with subtotal resection (p = 0.010) and high-grade meningiomas (p = 0.018).

Conclusions

Skull-base meningiomas with extracranial extensions were sporadic. According to the tumor location and communication way showed by the preoperative imaging, routine craniotomy or endoscopic endonasal approach with a reasonable skull-base repair strategy could be adopted to achieve the maximum tumor resection. Maximized resection, adjuvant radiotherapy, and low histological grade indicate a better prognosis.

Lateral Ventricular Meningiomas: Clinical Features, Radiological Findings and Long-Term Outcomes

Purpose

Lateral ventricle meningioma (LVM) is a rare type of intracranial meningioma, which has been rarely studied. It has different clinical features, imaging features, and long-term results from other locations. This study investigated the epidemiology, clinical characteristics and prognosis of LVM and comprehensively describes its characteristics.

Methods

This article analyzes the LVMs that were diagnosed pathologically in West China hospital between January 1, 2009 and July 1 2020. Demographic information, imaging characteristics and prognostic factors are discussed. Data analysis was performed using SPSS 23.0 and R version 3.5.3.

Results

We collected 7202 meningiomas and 195 LVMs (136 females; median age, 46 years; range, 5–81 years) were included in this study. Gross total resection was completed in 189 patients. The OS rate was 93.8%, and the recurrence rate was 5.2%. Multivariate regression analysis showed that sex (P = 0.01) and tumor size (P = 0.018) were related to WHO grade. Postoperative KPS (P = 0.003) was associated with OS. WHO grade (P = 0.025), extent of tumor resection (P < 0.001), and hospital day (P=0.028) were associated with recurrence.

Conclusion

LVMs require long-term follow-up, individualized treatment, and follow-up strategies to be formulated according to the relevant risk factors.

Nomograms for Predicting Overall Survival Among Patients with Craniopharyngiomas at Initial Diagnosis: A SEER Population-Based Analysis

Background

Craniopharyngiomas (CPs) are relatively rare benign tumor located in the central nervous system (CNS). This study investigates the related risk factors of survival of craniopharyngiomas and develops a simple but detailed method predicting prognosis based on the Surveillance, Epidemiology, and End Results (SEER) database in order to improve the clinic management of CPs.

Methods

Between 2004 and 2017, 1213 patients diagnosed with craniopharyngiomas registered at the program and were included in the SEER-21 registry database. Overall survival (OS) curves were plotted with the Kaplan–Meier method and significance was determined by Log rank test. Single- and multiple-factor regression analyses were made using Cox proportional hazards model to identify independent predictors related to OS. Subsequently, we developed a nomogram with those factors to predict 3-, 5- and 10-year OS of craniopharyngiomas patients.

Results

We identified 1213 patients with craniopharyngioma. The OS rates at 3, 5, and 10 years after diagnosis were 89.1%, 86.2%, and 83%. Age, ethnicity, tumor size and radiation therapy were confirmed to be predictors correlating with OS at initial diagnosis. In multivariate analysis, we found that younger age (P<0.001), smaller tumor size (P<0.001), white ethnicity (P<0.001) and radiation therapy (P=0.004) were the factors that remained significantly associated with better survival. A nomogram was successfully constructed and validated by ROC, calibration plots and C-index of 0.773 (95% CI, 0.708–0.838).

Conclusion

The well-calibrated nomogram is the first clinical prediction model for predicting the prognosis for patients with craniopharyngiomas at initial diagnosis. Our study indicates that the surgical effect is not clear. Younger white patients with radiotherapy have a better prognosis, and the gross total resection (GTR) was not effective in prolonging the OS of a patient compared to no surgery and subtotal resection (STR).

Pediatric meningiomas: 10-year experience with 39 patients

OBJECTIVE

The purpose of this article was to study the characteristics of pediatric meningiomas, including the clinical symptom, anatomic location, radiological finding, treatment approaches, and long-term surgical outcome as well as the possible influence factors of the prognosis.

METHODS

We retrospectively reviewed the medical records and follow-up data of 39 pediatric patients who were treated by operation for intracranial meningiomas in West-China Hospital between 2009-2019.

RESULTS

The incidence of pediatric meningioma was 0.74%. The mean age at surgery was 12.2 years and the ratio of male to female was 1.3. Three cases (7.7%) were related to neurofibromas. WHO grade I was found in 26 patients (66.7%) and higher grades was seen in 13 (33.3%). The most common location sites were convexity (35.9%). Gross total excision was achieved in 28 patients (71.8%). The mean follow-up period was 54.4 months. The recurrence rate was 41.9%. By survival analysis, only gross total resection (p = 0.028) was associated with favorable outcome.

CONCLUSION

Meningiomas in children are very rare and have a slight male predominance. Pediatric meningiomas are more commonly located in intraventricular and have higher incidence rate of high-grade than that in adults. Although being challenging, the gross total excision should be underdone to achieve a better prognosis.

Direct pathogen detection from swab samples using a new high-throughput sequencing technology

The detection of emerging infectious diseases has been a continuing concern, especially with the novel influenza A (H1N1) viral pandemic of 2009. In the present study, we validated a ‘second-generation’ parallel sequencing platform for viral detection in swab samples collected during recent influenza virus infections in Beijing. This operation yielded millions of valid reads per sample and resulted in an almost complete spectrum of nucleotide information. Importantly, novel A (H1N1) and seasonal A (H3N2) influenza virus-derived sequences were detected without prior knowledge or use of genetic information in advance, suggesting that this approach could be a valuable tool for diagnosing emerging infectious diseases.