Drone Model Classification Using Convolutional Neural Network Trained on Synthetic Data.

We present a convolutional neural network (CNN) that identifies drone models in real-life videos. The neural network is trained on synthetic images and tested on a real-life dataset of drone videos. To create the training and validation datasets, we show a method of generating synthetic drone images. Domain randomization is used to vary the simulation parameters such as model textures, background images, and orientation. Three common drone models are classified: DJI Phantom, DJI Mavic, and DJI Inspire. To test the performance of the neural network model, Anti-UAV, a real-life dataset of flying drones is used. The proposed method reduces the time-cost associated with manually labelling drones, and we prove that it is transferable to real-life videos. The CNN achieves an overall accuracy of 92.4%, a precision of 88.8%, a recall of 88.6%, and an f1 score of 88.7% when tested on the real-life dataset.

Simplified procedure of silymarin extraction from Silybum marianum L. Gaertner.

Silymarin, a mixture of flavonolignans exhibiting many pharmacological activities, is obtained from the fruits of milk thistle (Silybum marianum L. Gaertner). Due to the high lipid content in thistle fruits, the European Pharmacopoeia recommends a two-step process of its extraction. First, the fruits are defatted for 6 h, using n-hexane; second, silymarin is extracted with methanol for 5 more hours. The presented data show that this extremely long traditional Soxhlet extraction process can be shortened to a few minutes using pressurized liquid extraction (PLE). PLE also allows to eliminate the defatting stage required in the traditional procedure, thus simplifying the silymarin extraction procedure and preventing silymarin loss caused by defatting. The PLE recoveries obtained under the optimized extraction conditions are clearly better than the ones obtained by the Pharmacopoeia-recommended Soxhlet extraction procedure. The PLE yields of silychristin, silydianin, silybin A, silybin B, isosilybin A and isosilybin B in acetone are 3.3, 6.9, 3.3, 5.1, 2.6 and 1.5 mg/g of the non-defatted fruits, respectively. The 5-h Soxhlet extraction with methanol on defatted fruits gives only ∼72% of the silymarin amount obtained in 10 min PLE at 125°C.

[The impact of nitrates and mono-therapy and nitrates combined with angiotensin converting enzyme inhibitors on left ventricular remodeling and exercise capacity in patients after acute myocardial infarction]. / Porównanie wplywu nitratów w monoterapii oraz w skojarzeniu z inhibitorem konwertazy angiotensyny na remodelowanie lewej komory oraz wydolnosc wysilkowa u pacjentów po zawale serca.

The objective of this paper was to evaluate potential beneficial effects of combined treatment with slow-release nitrates and angiotensin converting enzyme inhibitors (ACE) on left ventricular remodeling and exercise capacity in patients after acute myocardial infarction. In this study, 141 patients (aged 34 to 74, mean 56.6 years) with sufficient circulation received combined treatment with 24 hour nitroglycerin infusion followed by oral nitrates (isosorbide mononitrate 50 mg OD) from day 2 day 42 after myocardial infarction and ACE inhibitor (captopril 25 mg BID or enalapril 5 mg BID versus placebo) from day 10 to day 42. On days 10 and 42, echocardiographic examination was carried out and recorded on an optical disc. Simultaneously, on the same days, the treadmill exercise test (modified Bruce protocol) was performed. In the echocardiographic study the left ventricular endodiastolic and endosystolic volumes (biplane Simpson formula), ejection fraction, left ventricular wall motion score and left ventricular mass index were analyzed. Treadmill test criteria, used in the study, included exercise duration time and workload (METS). For each patient the data obtained examination II and I were measured and the differences in their values were classified. The obtained results were analyzed with one-way and three-way ANOVA test. A Kruskal-Wallis test was also used in one variable analysis. Results were analyzed after repartition of patients into groups according type of treatment (angiotensin converting ing enzyme inhibitor or placebo), infarct location (anterior or inferior wall) and enzyme level (CPK < 2000 IU/L or CPK > 2000 IU/L). A p value < 0.1 was considered statistically significant. In a single factor analysis ANOVA proved that the patients treated with nitrates and captopril showed greater improvement in exercise capacity (in METS) than patients treated with enalapril or placebo (+1.26 captopril, +0.2 enalapril and +0.29 placebo, p = 0.043). In addition, a decrease in left ventricular mass index was evident only in patients treated with angiotensin converting enzyme inhibitor (placebo +7.37 gm/m2, captopril -12.17 gm/m2, enalapril -10.14 gm/m2, p = 0.0032). The triple factor analysis ANOVA test revealed that the change in endodiastolic left ventricular volume depends on combination of three factors: infarct location, type of treatment and level of cardiac enzymes (p = 0.009). A decrease in left ventricular endodiastolic volume between day 42 and 10 was observed only in patients with inferior wall infarct and CPK level < 2000 IU/L, irrespective of treatment type and in patients with inferior wall infarct and CPK level > 2000 IU/L treated with angiotensin enzyme inhibitor. We noticed also that heart failure, considered as contraindication to randomization, was in addition the most frequent (up to day 10) cause for study termination and initiation of treatment with angiotensin enzyme inhibitor.