Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Deep Learning-based Assessment of Internal Carotid Artery Anatomy to Predict Difficult Intracranial Access in Endovascular Recanalization of Acute Ischemic Stroke.

In Clinical neuroradiology

BACKGROUND : Endovascular thrombectomy (EVT) duration is an important predictor for neurological outcome. Recently it was shown that an angle of ≤ 90° of the internal carotid artery (ICA) is predictive for longer EVT duration. As manual angle measurement is not trivial and time-consuming, deep learning (DL) could help identifying difficult EVT cases in advance.

METHODS : We included 379 CT angiographies (CTA) of patients who underwent EVT between January 2016 and December 2020. Manual segmentation of 121 CTAs was performed for the aortic arch, common carotid artery (CCA) and ICA. These were used to train a nnUNet. The remaining 258 CTAs were segmented using the trained nnUNet with manual verification afterwards. Angles of left and right ICAs were measured resulting in two classes: acute angle ≤ 90° and > 90°. The segmentations together with angle measurements were used to train a convolutional neural network (CNN) determining the ICA angle. The performance was evaluated using Dice scores. The classification was evaluated using AUC and accuracy. Associations of ICA angle and procedural times was explored using median and Whitney‑U test.

RESULTS : Median EVT duration for cases with ICA angle > 90° was 48 min and with ≤ 90° was 64 min (p = 0.001). Segmentation evaluation showed Dice scores of 0.94 for the aorta and 0.86 for CCA/ICA, respectively. Evaluation of ICA angle determination resulted in an AUC of 0.92 and accuracy of 0.85.

CONCLUSION : The association between ICA angle and EVT duration could be verified and a DL-based method for semi-automatic assessment with the potential for full automation was developed. More anatomical features of interest could be examined in a similar fashion.

Nageler Gregor, Gergel Ingmar, Fangerau Markus, Breckwoldt Michael, Seker Fatih, Bendszus Martin, Möhlenbruch Markus, Neuberger Ulf

2023-Mar-16

Convolutional neural network (CNN), Machine learning, Mechanical thrombectomy, Tortuosity, nnUNet

General General

Metabolomics integrated with machine learning to discriminate the geographic origin of Rougui Wuyi rock tea.

In NPJ science of food

The geographic origin of agri-food products contributes greatly to their quality and market value. Here, we developed a robust method combining metabolomics and machine learning (ML) to authenticate the geographic origin of Wuyi rock tea, a premium oolong tea. The volatiles of 333 tea samples (174 from the core region and 159 from the non-core region) were profiled using gas chromatography time-of-flight mass spectrometry and a series of ML algorithms were tested. Wuyi rock tea from the two regions featured distinct aroma profiles. Multilayer Perceptron achieved the best performance with an average accuracy of 92.7% on the training data using 176 volatile features. The model was benchmarked with two independent test sets, showing over 90% accuracy. Gradient Boosting algorithm yielded the best accuracy (89.6%) when using only 30 volatile features. The proposed methodology holds great promise for its broader applications in identifying the geographic origins of other valuable agri-food products.

Peng Yifei, Zheng Chao, Guo Shuang, Gao Fuquan, Wang Xiaxia, Du Zhenghua, Gao Feng, Su Feng, Zhang Wenjing, Yu Xueling, Liu Guoying, Liu Baoshun, Wu Chengjian, Sun Yun, Yang Zhenbiao, Hao Zhilong, Yu Xiaomin

2023-Mar-16

Internal Medicine Internal Medicine

Adapted clustering method for generic analysis of histological fibrosis staining as an open source tool.

In Scientific reports ; h5-index 158.0

Pathological remodeling of the extracellular matrix is a hallmark of cardiovascular disease. Abnormal fibrosis causes cardiac dysfunction by reducing ejection fraction and impairing electrical conductance, leading to arrhythmias. Hence, accurate quantification of fibrosis deposition in histological sections is of extreme importance for preclinical and clinical studies. Current automatic tools do not perform well under variant conditions. Moreover, users do not have the option to evaluate data from staining methods of their choice according to their purpose. To overcome these challenges, we underline a novel machine learning-based tool (FibroSoft) and we show its feasibility in a model of cardiac hypertrophy and heart failure in mice. Our results demonstrate that FibroSoft can identify fibrosis in diseased myocardium and the obtained results are user-independent. In addition, the results acquired using our software strongly correlate to those obtained by Western blot analysis of collagen 1 expression. Additionally, we could show that this method can be used for Masson's Trichrome and Picosirius Red stained histological images. The evaluation of our method also indicates that it can be used for any particular histology segmentation and quantification. In conclusion, our approach provides a powerful example of the feasibility of machine learning strategies to enable automatic analysis of histological images.

Remes Anca, Noormalal Marie, Schmiedel Nesrin, Frey Norbert, Frank Derk, Müller Oliver J, Graf Markus

2023-Mar-16

General General

Artificial intelligence-based optimization for chitosan nanoparticles biosynthesis, characterization and in‑vitro assessment of its anti-biofilm potentiality.

In Scientific reports ; h5-index 158.0

Chitosan nanoparticles (CNPs) are promising biopolymeric nanoparticles with excellent physicochemical, antimicrobial, and biological properties. CNPs have a wide range of applications due to their unique characteristics, including plant growth promotion and protection, drug delivery, antimicrobials, and encapsulation. The current study describes an alternative, biologically-based strategy for CNPs biosynthesis using Olea europaea leaves extract. Face centered central composite design (FCCCD), with 50 experiments was used for optimization of CNPs biosynthesis. The artificial neural network (ANN) was employed for analyzing, validating, and predicting CNPs biosynthesis using Olea europaea leaves extract. Using the desirability function, the optimum conditions for maximum CNPs biosynthesis were determined theoretically and verified experimentally. The highest experimental yield of CNPs (21.15 mg CNPs/mL) was obtained using chitosan solution of 1%, leaves extract solution of 100%, initial pH 4.47, and incubation time of 60 min at 53.83°C. The SEM and TEM images revealed that CNPs had a spherical form and varied in size between 6.91 and 11.14 nm. X-ray diffraction demonstrates the crystalline nature of CNPs. The surface of the CNPs is positively charged, having a Zeta potential of 33.1 mV. FTIR analysis revealed various functional groups including C-H, C-O, CONH2, NH2, C-OH and C-O-C. The thermogravimetric investigation indicated that CNPs are thermally stable. The CNPs were able to suppress biofilm formation by P. aeruginosa, S. aureus and C. albicans at concentrations ranging from 10 to 1500 µg/mL in a dose-dependent manner. Inhibition of biofilm formation was associated with suppression of metabolic activity, protein/exopolysaccharide moieties, and hydrophobicity of biofilm encased cells (r ˃ 0.9, P = 0.00). Due to their small size, in the range of 6.91 to 11.14 nm, CNPs produced using Olea europaea leaves extract are promising for applications in the medical and pharmaceutical industries, in addition to their potential application in controlling multidrug-resistant microorganisms, especially those associated with post COVID-19 pneumonia in immunosuppressed patients.

El-Naggar Noura El-Ahmady, Dalal Shimaa R, Zweil Amal M, Eltarahony Marwa

2023-Mar-16

General General

Mapping of Phragmites in estuarine wetlands using high-resolution aerial imagery.

In Environmental monitoring and assessment

Phragmites australis is a widespread invasive plant species in the USA that greatly impacts estuarine wetlands by creating dense patches and outcompeting other plants. The invasion of Phragmites into wetland ecosystems is known to decrease biodiversity, destroy the habitat of threatened and endangered bird species, and alter biogeochemistry. While the impact of Phragmites is known, the spatial extent of this species is challenging to document due to its fragmented occurrence. Using high-resolution imagery from the National Agriculture Imagery Program (NAIP) from 2017, we evaluated a geospatial method of mapping the spatial extent of Phragmites across the state of DE. Normalized difference vegetation index (NDVI) and principal component analysis (PCA) bands are generated from the NAIP data and used as inputs in a random forest classifier to achieve a high overall accuracy for the Phragmites classification of around 95%. The classified gridded dataset has a spatial resolution of 1 m and documents the spatial distribution of Phragmites throughout the state's estuarine wetlands (around 11%). Such detailed classification could aid in monitoring the spread of this invasive species over space and time and would inform the decision-making process for landscape managers.

Walter Matthew, Mondal Pinki

2023-Mar-17

Classification, Invasive, Machine learning, Phragmites, Vegetation

Pathology Pathology

Challenges Posed by the Banff Classification: Diagnosis and Treatment of Chronic active T-cell mediated Rejection.

In Nephron

The three primary sites of acute T-cell mediated rejection (TCMR) in transplanted kidneys are the tubular epithelial cells, interstitum, and the vascular endothelial cells. The pathology of acute lesions is characterized by inflammatory cell infiltration; the final diagnosis suggested by the Banff 2019 classification is guided by grading of tubulitis (the t-score), interstitial inflammation (the i score), and endarteritis (the v score). Consistent major issues when using the Banff classification are the etiological classifications of interstitial fibrosis and tubular atrophy (IFTA). From 2015 to 2019, technological advances (i.e., genetic analysis in paraffin sections), increased our understanding of IFTA status in patients with smoldering acute TCMR and the roles played by inflammatory cell infiltration (the i-IFTA score) and tubulitis (the t-IFTA score) in IFTA. These two scores were introduced when establishing the diagnostic criteria for chronic active TCMR. Despite the increase in complexity and the lack of a consensus treatment for chronic active TCMR, the Banff classification may evolve as new techniques (i.e., genetic analysis in paraffin sections and deep learning of renal pathology), are introduced. The Banff conference proceeded as follows. First, lesions were defined. Next, working groups were established to better understand the lesions and to derive better classification methods. Finally, the new Banff classification was developed. This approach will continue to evolve; the Banff classification will become a very useful diagnostic standard. This paper overviews the history of TCMR diagnosis using the Banff classification, and the clinical importance, treatment, and prospects for acute and chronic active TCMR.

Yamamoto Izumi, Kawabe Mayuko, Hayashi Ayaka, Kobayashi Akimitsu, Yamamoto Hiroyasu, Yokoo Takashi

2023-Mar-16