Deep Learning-Based Differential Diagnosis of Follicular Thyroid Tumors Using Histopathological Images.

Deep learning systems (DLSs) have been developed for the histopathological assessment of various types of tumors, but none are suitable for differential diagnosis between follicular thyroid carcinoma (FTC) and follicular adenoma (FA). Furthermore, whether DLSs can identify the malignant characteristics of thyroid tumors based only on random views of tumor tissue histology has not been evaluated. In this study, we developed DLSs able to differentiate between FTC and FA based on 3 types of convolutional neural network architecture: EfficientNet, VGG16, and ResNet50. The performance of all 3 DLSs was excellent (area under the receiver operating characteristic curve = 0.91 ± 0.04; F1 score = 0.82 ± 0.06). Visual explanations using gradient-weighted class activation mapping suggested that the diagnosis of both FTC and FA was largely dependent on nuclear features. The DLSs were then trained with FTC images and linked information (presence or absence of recurrence within 10 years, vascular invasion, and wide capsular invasion). The ability of the DLSs to diagnose these characteristics was then determined. The results showed that, based on the random views of histology, the DLSs could predict the risk of FTC recurrence, vascular invasion, and wide capsular invasion with a certain level of accuracy (area under the receiver operating characteristic curve = 0.67 ± 0.13, 0.62 ± 0.11, and 0.65 ± 0.09, respectively). Further improvement of our DLSs could lead to the establishment of automated differential diagnosis systems requiring only biopsy specimens.

Acute myocardial infarction due to traumatic coronary artery dissection: conflicts with thrombotic and bleeding pathologies.

Traumatic coronary artery dissection is a rare and challenging condition, requiring treatments for both thrombus and haemorrhage. We present a unique case of traumatic coronary artery dissection in which intra-aortic balloon pumping without heparin usage was useful in the acute traumatic phase of haemorrhagic shock with a contraindication to anti-thrombotic therapy.

Requirement of glycosylation machinery in TLR responses revealed by CRISPR/Cas9 screening.

The Toll family of receptors sense microbial products and activate a defense response. The molecular machinery required for the TLR response is not yet fully understood. In the present study, we used a clustered, regularly interspaced, short palindromic repeats (CRISPR)/CAS9 screening system to study TLR responses. We employed a cell line expressing TLR with an NF-κB-driven GFP reporter. The cell line was transduced with a guide RNA (gRNA) library and stimulated with TLR ligands. The cells impaired in GFP induction were sorted, and gRNAs were sequenced. Identified genes were ranked according to the count of sequence reads and the number of gRNA target sites. The screening system worked correctly, as molecules that were already known to be required for the TLR response were identified by the screening. Furthermore, this system revealed that the oligosaccharide transferase complex (OSTC) mediating co-translational glycosylation was required for TLR5, 7 and 9 responses. Protein expression of TLR5, but not an irrelevant molecule (CD44), was abolished by the lack of OSTC, suggesting the essential role of glycosylation in TLR5 protein stability. These results demonstrate that the screening system established here is able to reveal molecular mechanisms underlying the TLR response.

Growth patterns in Onychophora (velvet worms): lack of a localised posterior proliferation zone.

BACKGROUND: During embryonic development of segmented animals, body segments are thought to arise from the so-called "posterior growth zone" and the occurrence of this "zone" has been used to support the homology of segmentation between arthropods, annelids, and vertebrates. However, the term "posterior growth zone" is used ambiguously in the literature, mostly referring to a region of increased proliferation at the posterior end of the embryo. To determine whether such a localised posterior proliferation zone is an ancestral feature of Panarthropoda (Onychophora + Tardigrada + Arthropoda), we examined cell division patterns in embryos of Onychophora. RESULTS: Using in vivo incorporation of the DNA replication marker BrdU (5-bromo-2'-deoxyuridine) and anti-phospho-histone H3 immunolabelling, we found that a localised posterior region of proliferating cells does not occur at any developmental stage in onychophoran embryos. This contrasts with a localised pattern of cell divisions at the posterior end of annelid embryos, which we used as a positive control. Based on our data, we present a mathematical model, which challenges the paradigm that a localised posterior proliferation zone is necessary for segment patterning in short germ developing arthropods. CONCLUSIONS: Our findings suggest that a posterior proliferation zone was absent in the last common ancestor of Onychophora and Arthropoda. By comparing our data from Onychophora with those from annelids, arthropods, and chordates, we suggest that the occurrence of a "posterior growth zone" currently cannot be used to support the homology of segmentation between these three animal groups.

Perianth bottom-specific blue color development in Tulip cv. Murasakizuisho requires ferric ions.

The entire flower of Tulipa gesneriana cv. Murasakizuisho is purple, except the bottom, which is blue. To elucidate the mechanism of the different color development in the same petal, we prepared protoplasts from the purple and blue epidermal regions and measured the flavonoid composition by HPLC, the vacuolar pH by a proton-selective microelectrode, and element contents by the inductively coupled plasma (ICP) method. Chemical analyses revealed that the anthocyanin and flavonol compositions in both purple and blue colored protoplasts were the same; delphinidin 3-O-rutinoside (1) and major three flavonol glycosides, manghaslin (2), rutin (3) and mauritianin (4). The vacuolar pH values of the purple and blue protoplasts were 5.5 and 5.6, respectively, without any significant difference. However, the Fe(3+) content in the blue protoplast was approximately 9.5 mM, which was 25 times higher than that in the purple protoplasts. We could reproduce the purple solution by mixing 1 with two equimolar concentrations of flavonol with lambda(vismax) = 539 nm, which was identical to that of the purple protoplasts. Furthermore, addition of Fe(3+) to the mixture of 1-4 gave the blue solution with lambda(vismax) = 615 nm identical to that of the blue protoplasts. We have established that Fe(3+) is essential for blue color development in the tulip.

Differential expression of the nitrite reductase gene family in tobacco as revealed by quantitative competitive RT-PCR.

Tobacco (Nicotiana tabacum L. cv. Xanthi XHFD8) possesses four nitrite reductase (NiR) genes: nii1, nii2, nii3, and nii4. Their differential expression in leaves and roots was investigated by quantitative competitive RT-PCR using gene-specific primer pairs. These results appear to contradict existing views on the expression of these NiR genes: (i) the mRNA of each of the four NiR genes was distinguishable both in leaves and roots; (ii) nitrate treatment increased nii1 and nii3 mRNA in leaves and roots by at least 4-fold (at least 5-fold in nii2 and nii4 mRNA); and (iii) the steady-state levels of nii1 and nii3 mRNA were almost the same in leaves (6-7 x 10(5) and about 3 x 10(6) copies microg(-1) of total RNA before and after nitrate treatment, respectively) and in roots (3-4 x 10(4) and 3-6 x 10(5) copies microg(-1) of total RNA before and after nitrate treatment, respectively). Very similar relationships were obtained for the steady-state levels of nii2 and nii4 mRNA in roots (2-4 x 10(5) and 8 x 10(6) copies microg(-1) of total RNA before and after nitrate treatment, respectively), and in leaves (5-9 x 10(4) and 4 x 10(5) copies microg(-1) of total RNA before and after nitrate treatment, respectively). These results demonstrate that nii1 and nii3 transcripts are a dominating, but not exclusive, NiR mRNA in leaves, and the same is true for nii2 and nii4 transcripts in roots.