Automated Mapping of Land Cover Type within International Heterogenous Landscapes Using Sentinel-2 Imagery with Ancillary Geospatial Data.

A near-global framework for automated training data generation and land cover classification using shallow machine learning with low-density time series imagery does not exist. This study presents a methodology to map nine-class, six-class, and five-class land cover using two dates (winter and non-winter) of a Sentinel-2 granule across seven international sites. The approach uses a series of spectral, textural, and distance decision functions combined with modified ancillary layers (such as global impervious surface and global tree cover) to create binary masks from which to generate a balanced set of training data applied to a random forest classifier. For the land cover masks, stepwise threshold adjustments were applied to reflectance, spectral index values, and Euclidean distance layers, with 62 combinations evaluated. Global (all seven scenes) and regional (arid, tropics, and temperate) adaptive thresholds were computed. An annual 95th and 5th percentile NDVI composite was used to provide temporal corrections to the decision functions, and these corrections were compared against the original model. The accuracy assessment found that the regional adaptive thresholds for both the two-date land cover and the temporally corrected land cover could accurately map land cover type within nine-class (68.4% vs. 73.1%), six-class (79.8% vs. 82.8%), and five-class (80.1% vs. 85.1%) schemes. Lastly, the five-class and six-class models were compared with a manually labeled deep learning model (Esri), where they performed with similar accuracies (five classes: Esri 80.0 ± 3.4%, region corrected 85.1 ± 2.9%). The results highlight not only performance in line with an intensive deep learning approach, but also that reasonably accurate models can be created without a full annual time series of imagery.

Correction: Bolocan et al. Convolutional Neural Network Model for Segmentation and Classification of Clear Cell Renal Cell Carcinoma Based on Multiphase CT Images. J. Imaging 2023, 9, 280.

In the original publication [...].

Convolutional Neural Network Model for Segmentation and Classification of Clear Cell Renal Cell Carcinoma Based on Multiphase CT Images.

(1) Background: Computed tomography (CT) imaging challenges in diagnosing renal cell carcinoma (RCC) include distinguishing malignant from benign tissues and determining the likely subtype. The goal is to show the algorithm's ability to improve renal cell carcinoma identification and treatment, improving patient outcomes. (2) Methods: This study uses the European Deep-Health toolkit's Convolutional Neural Network with ECVL, (European Computer Vision Library), and EDDL, (European Distributed Deep Learning Library). Image segmentation utilized U-net architecture and classification with resnet101. The model's clinical efficiency was assessed utilizing kidney, tumor, Dice score, and renal cell carcinoma categorization quality. (3) Results: The raw dataset contains 457 healthy right kidneys, 456 healthy left kidneys, 76 pathological right kidneys, and 84 pathological left kidneys. Preparing raw data for analysis was crucial to algorithm implementation. Kidney segmentation performance was 0.84, and tumor segmentation mean Dice score was 0.675 for the suggested model. Renal cell carcinoma classification was 0.885 accurate. (4) Conclusion and key findings: The present study focused on analyzing data from both healthy patients and diseased renal patients, with a particular emphasis on data processing. The method achieved a kidney segmentation accuracy of 0.84 and mean Dice scores of 0.675 for tumor segmentation. The system performed well in classifying renal cell carcinoma, achieving an accuracy of 0.885, results which indicates that the technique has the potential to improve the diagnosis of kidney pathology.

Motion of Adsorbed Nano-Particles on Azobenzene Containing Polymer Films.

We demonstrate in situ recorded motion of nano-objects adsorbed on a photosensitive polymer film. The motion is induced by a mass transport of the underlying photoresponsive polymer material occurring during irradiation with interference pattern. The polymer film contains azobenzene molecules that undergo reversible photoisomerization reaction from trans- to cis-conformation. Through a multi-scale chain of physico-chemical processes, this finally results in the macro-deformations of the film due to the changing elastic properties of polymer. The topographical deformation of the polymer surface is sensitive to a local distribution of the electrical field vector that allows for the generation of dynamic changes in the surface topography during irradiation with different light interference patterns. Polymer film deformation together with the motion of the adsorbed nano-particles are recorded using a homemade set-up combining an optical part for the generation of interference patterns and an atomic force microscope for acquiring the surface deformation. The particles undergo either translational or rotational motion. The direction of particle motion is towards the topography minima and opposite to the mass transport within the polymer film. The ability to relocate particles by photo-induced dynamic topography fluctuation offers a way for a non-contact simultaneous manipulation of a large number of adsorbed particles just in air at ambient conditions.

A comparative study of photoinduced deformation in azobenzene containing polymer films.

In this paper two groups supporting different views on the mechanism of light induced polymer deformation argue about the respective underlying theoretical conceptions, in order to bring this interesting debate to the attention of the scientific community. The group of Prof. Nicolae Hurduc supports the model claiming that the cyclic isomerization of azobenzenes may cause an athermal transition of the glassy azobenzene containing polymer into a fluid state, the so-called photo-fluidization concept. This concept is quite convenient for an intuitive understanding of the deformation process as an anisotropic flow of the polymer material. The group of Prof. Svetlana Santer supports the re-orientational model where the mass-transport of the polymer material accomplished during polymer deformation is stated to be generated by the light-induced re-orientation of the azobenzene side chains and as a consequence of the polymer backbone that in turn results in local mechanical stress, which is enough to irreversibly deform an azobenzene containing material even in the glassy state. For the debate we chose three polymers differing in the glass transition temperature, 32 °C, 87 °C and 95 °C, representing extreme cases of flexible and rigid materials. Polymer film deformation occurring during irradiation with different interference patterns is recorded using a homemade set-up combining an optical part for the generation of interference patterns and an atomic force microscope for acquiring the kinetics of film deformation. We also demonstrated the unique behaviour of azobenzene containing polymeric films to switch the topography in situ and reversibly by changing the irradiation conditions. We discuss the results of reversible deformation of three polymers induced by irradiation with intensity (IIP) and polarization (PIP) interference patterns, and the light of homogeneous intensity in terms of two approaches: the re-orientational and the photo-fluidization concepts. Both agree in that the formation of opto-mechanically induced stresses is a necessary prerequisite for the process of deformation. Using this argument, the deformation process can be characterized either as a flow or mass transport.

High prevalence and breast cancer predisposing role of the BLM c.1642 C>T (Q548X) mutation in Russia.

The BLM gene belongs to the RecQ helicase family and has been implicated in the maintenance of genomic stability. Its homozygous germline inactivation causes Bloom syndrome, a severe genetic disorder characterized by growth retardation, impaired fertility and highly elevated cancer risk. We hypothesized that BLM is a candidate gene for breast cancer (BC) predisposition. Sequencing of its entire coding region in 95 genetically enriched Russian BC patients identified two heterozygous carriers of the c.1642 C>T (Q548X) mutation. The extended study revealed this allele in 17/1,498 (1.1%) BC cases vs. 2/1,093 (0.2%) healthy women (p = 0.004). There was a suggestion that BLM mutations were more common in patients reporting first-degree family history of BC (6/251 (2.4%) vs. 11/1,247 (0.9%), p = 0.05), early-onset cases (12/762 (1.6%) vs. 5/736 (0.7%), p = 0.14) and women with bilateral appearance of the disease (2/122 (1.6%) vs. 15/1376 (1.1%), p = 0.64). None of the BLM-associated BC exhibited somatic loss of heterozygosity at the BLM gene locus. This study demonstrates that BLM Q548X allele is recurrent in Slavic subjects and may be associated with BC risk.

Large family with both parents affected by distinct BRCA1 mutations: implications for genetic testing.

Although the probability of both parents being affected by BRCA1 mutations is not negligible, such families have not been systematically described in the literature. Here we present a large breast-ovarian cancer family, where 3 sisters and 1 half-sister inherited maternal BRCA1 5382insC mutation while the remaining 2 sisters carried paternal BRCA1 1629delC allele. No BRCA1 homozygous mutations has been detected, that is consistent with the data on lethality of BRCA1 knockout mice. This report exemplifies that the identification of a single cancer-predisposing mutation within the index patient may not be sufficient in some circumstances. Ideally, all family members affected by breast or ovarian tumor disease have to be subjected to the DNA testing, and failure to detect the mutation in any of them calls for the search of the second cancer-associated allele.

Founder mutations in early-onset, familial and bilateral breast cancer patients from Russia.

Previous studies indicate that founder mutations may play a noticeable role in breast cancer (BC) predisposition in Russia. Here we performed a systematic analysis of eight recurrent mutations in 302 BC cases (St.-Petersburg, Russia), which were selected due to the presence of clinical indicators of hereditary disease (bilaterality and/or early onset (< or =40 years) and/or family history). BC-associated alleles were revealed in 46 (15.2%) women. BRCA1 5382insC mutation was detected in 29 (9.6%) patients, CHEK2 1100delC in 9 (3.0%), BRCA1 4153delA in 3 (1.0%), CHEK2 IVS2+1G>A in 2 (0.7%), and BRCA1 185delAG, BRCA2 6174delT and NBS1 657del5 in 1 (0.3%) patient each. No cases with BRCA1 300T>G (C61G) mutation was identified. The obtained data suggest that a significant fraction of hereditary BC cases in Russia can be diagnosed using only a limited number of simple PCR tests.

High frequency of BRCA1 5382insC mutation in Russian breast cancer patients.

BRCA1 5382insC variant was repeatedly detected in Jewish breast cancer (BC) families residing in USA and Israel as well as in non-Jewish familial BC patients from Poland, Latvia, Hungary, Russia and some other European countries. However, the distribution of BRCA1 5382insC mutation in unselected BC cases vs. controls has been systematically investigated mainly in Ashkenazi Jews. Here we applied a case-control study design in order to evaluate the impact of BRCA1 5382insC allele on BC incidence in St Petersburg, Russia. High frequency of the BRCA1 5382insC allele was detected in a group of bilateral breast cancer patients (10.4%; 15/144). Randomly selected unilateral BC cases demonstrated noticeable occurrence of BRCA1 5382insC mutation as well (3.7%; 32/857), with evident excess of the carriers in the early-onset (40 years) category (6.1%; 6/99) and in patients reporting breast and/or ovarian tumours in first-degree relatives (11.3%; 11/97). Strikingly, none of 478 middle-aged controls and 344 elderly tumour-free women carried the 5382insC variant. The presented data confirm a noticeable contribution of BRCA1 5382insC mutation in BC development in Russia, that may justify an extended BRCA1 5382insC testing within this population.

NBS1 657del5 mutation may contribute only to a limited fraction of breast cancer cases in Russia.

The gene for Nijmegen chromosomal breakage syndrome (NBS1) plays a role in a variety of processes protecting chromosomal stability. Recently, it was suggested in a Polish case-control study that the founder hypomorphic mutation in NBS1, 657del5, which occurs in approximately 0.5% of Slavic subjects, may be associated with an increased risk of breast cancer (BC). We attempted to validate these findings in Russian subjects, who are also of Slavic descent. Heterozygous carriers for the 657del5 mutation were detected in 2 of 173 (1.16%) bilateral breast cancer cases, 5 of 700 (0.71%) unilateral breast cancer patients, 2 of 348 (0.57%) healthy middle-aged females and in 0 of 344 elderly tumor-free women. The difference between the "extreme" cohorts, i.e., biBC patients vs. elderly controls, approached the formal limit of statistic significance (p=0.046). LOH at NBS1 was detected in only 3 of 5 available breast tumors from NBS1 657del5-carriers. In 2 of these tumors, the loss involved the mutant NBS1-allele. Overall, our data suggest that the NBS1 657del5 allele may contribute only to a limited fraction of breast cancer cases in Russia.