CODD: A benchmark dataset for the automated sorting of construction and demolition waste.

This study presents the Construction and Demolition Waste Object Detection Dataset (CODD), a benchmark dataset specifically curated for the training of object detection models and the full-scale implementation of automated sorting of Construction and Demolition Waste (CDW). The CODD encompasses a comprehensive range of CDW scenarios, capturing a diverse array of debris and waste materials frequently encountered in real-world construction and demolition sites. A noteworthy feature of the presented study is the ongoing collaborative nature of the dataset, which invites contributions from the scientific community, ensuring its perpetual improvement and adaptability to emerging research and practical requirements. Building upon the benchmark dataset, an advanced object detection model based on the latest bounding box and instance segmentation YOLOV8 architecture is developed to establish a baseline performance for future comparisons. The CODD benchmark dataset, along with the baseline model, provides a reliable reference for comprehensive comparisons and objective assessments of future models, contributing to progressive advancements and collaborative research in the field.

Reconciling a national methane emission inventory with in-situ measurements.

Reconciling top-down and bottom-up country-level greenhouse gas emission estimates remains a key challenge in the MRV (Monitoring, Reporting, Verification) paradigm. Here we propose to independently quantify cumulative emissions from a significant number of methane (CH4) emitters at national level and derive robust constraints for the national inventory. Methane emissions in Cyprus, an insular country, stem primarily from waste and agricultural activities. We performed 24 intensive survey days of mobile measurements of CH4 from October 2020 to September 2021 at emission 'hotspots' in Cyprus accounting together for about 28 % of national CH4 emissions. The surveyed areas include a large active landfill (Koshi, 8 % of total emissions), a large closed landfill (Kotsiatis, 18 %), and a concentrated cattle farm area (Aradippou, 2 %). Emission rates for each site were estimated using repeated downwind transects and a Gaussian plume dispersion model. The calculated methane emissions from landfills of Koshi and Kotsiatis (25.9 ± 6.4 Gg yr-1) and enteric fermentation of cattle (10.4 ± 4.4 Gg yr-1) were about 129 % and 40 % larger, respectively than the bottom-up sectorial annual estimates used in the national UNFCCC inventory. The parametrization of the Gaussian plume model dominates the uncertainty in our method, with a typical 21 % uncertainty. Seasonal variations have little influence on the results. We show that using an ensemble of in situ measurements targeting representative methane emission hotspots with consistent temporal and spatial coverage can contribute to the monitoring and validation of national bottom-up emission inventories.

Real-time construction demolition waste detection using state-of-the-art deep learning methods; single-stage vs two-stage detectors.

Central to the development of a successful waste sorting robot lies an accurate and fast object detection system. This study assesses the performance of the most representative deep-learning models for the real-time localisation and classification of Construction and Demolition Waste (CDW). For the investigation, both single-stage (SSD, YOLO) and two-stage (Faster-RCNN) detector architectures coupled with various backbone feature extractors (ResNet, MobileNetV2, efficientDet) were considered. A total of 18 models of variable depth were trained and tested on the first openly accessible CDW dataset developed by the authors of this study. This dataset consists of images of 6600 samples of CDW belonging to three object categories: brick, concrete, and tile. For an in-depth examination of the performance of the developed models under working conditions, two testing datasets containing normally and heavily stacked and adhered samples of CDW were developed. A comprehensive comparison between the different models yields that the latest version of the YOLO series (YoloV7) attains the best accuracy (mAP50:95 ≈ 70%) at the highest inference speed (<30 ms), while also exhibiting enough precision to deal with severely stacked and adhered samples of CDW. Additionally, it was observed that despite the rising popularity of single-stage detectors, apart from YoloV7, Faster-RCNN models remain the most robust in terms of exhibiting the least mAP fluctuations over the testing datasets considered.

A model for determining cardiac mitochondrial substrate utilisation using stable 13C-labelled metabolites.

INTRODUCTION: Relative oxidation of different metabolic substrates in the heart varies both physiologically and pathologically, in order to meet metabolic demands under different circumstances. 13C labelled substrates have become a key tool for studying substrate use-yet an accurate model is required to analyse the complex data produced as these substrates become incorporated into the Krebs cycle. OBJECTIVES: We aimed to generate a network model for the quantitative analysis of Krebs cycle intermediate isotopologue distributions measured by mass spectrometry, to determine the 13C labelled proportion of acetyl-CoA entering the Krebs cycle. METHODS: A model was generated, and validated ex vivo using isotopic distributions measured from isolated hearts perfused with buffer containing 11 mM glucose in total, with varying fractions of universally labelled with 13C. The model was then employed to determine the relative oxidation of glucose and triacylglycerol by hearts perfused with 11 mM glucose and 0.4 mM equivalent Intralipid (a triacylglycerol mixture). RESULTS: The contribution of glucose to Krebs cycle oxidation was measured to be 79.1 ± 0.9%, independent of the fraction of buffer glucose which was U-13C labelled, or of which Krebs cycle intermediate was assessed. In the presence of Intralipid, glucose and triglyceride were determined to contribute 58 ± 3.6% and 35.6 ± 0.8% of acetyl-CoA entering the Krebs cycle, respectively. CONCLUSION: These results demonstrate the accuracy of a functional model of Krebs cycle metabolism, which can allow quantitative determination of the effects of therapeutics and pathology on cardiac substrate metabolism.