Attention feature fusion methodology with additional constraint for ovarian lesion diagnosis on magnetic resonance images.

PURPOSE: It is challenging for radiologists and gynecologists to identify the type of ovarian lesions by reading magnetic resonance (MR) images. Recently developed convolutional neural networks (CNNs) have made great progress in computer vision, but their architectures still need modification if they are used in processing medical images. This study aims to improve the feature extraction capability of CNNs, thus promoting the diagnostic performance in discriminating between benign and malignant ovarian lesions. METHODS: We introduce a feature fusion architecture and insert the attention models in the neural network. The features extracted from different middle layers are integrated with reoptimized spatial and channel weights. We add a loss function to constrain the additional probability vector generated from the integrated features, thus guiding the middle layers to emphasize useful information. We analyzed 159 lesions imaged by dynamic contrast-enhanced MR imaging (DCE-MRI), including 73 benign lesions and 86 malignant lesions. Senior radiologists selected and labeled the tumor regions based on the pathology reports. Then, the tumor regions were cropped into 7494 nonoverlapping image patches for training and testing. The type of a single tumor was determined by the average probability scores of the image patches belonging to it. RESULTS: We implemented fivefold cross-validation to characterize our proposed method, and the distribution of performance matrics was reported. For all the test image patches, the average accuracy of our method is 70.5% with an average area under the curve (AUC) of 0.785, while the baseline is 69.4% and 0.773, and for the diagnosis of single tumors, our model achieved an average accuracy of 82.4% and average AUC of 0.916, which were better than the baseline (81.8% and 0.899). Moreover, we evaluated the performance of our proposed method utilizing different CNN backbones and different attention mechanisms. CONCLUSIONS: The texture features extracted from different middle layers are crucial for ovarian lesion diagnosis. Our proposed method can enhance the feature extraction capabilities of different layers of the network, thereby improving diagnostic performance.

Deep learning with a convolutional neural network model to differentiate renal parenchymal tumors: a preliminary study.

PURPOSE: With advancements in medical imaging, more renal tumors are detected early, but it remains a challenge for radiologists to accurately distinguish subtypes of renal parenchymal tumors. We aimed to establish a novel deep convolutional neural network (CNN) model and investigate its effect on identifying subtypes of renal parenchymal tumors in T2-weighted fat saturation sequence magnetic resonance (MR) images. METHODS: This retrospective study included 199 patients with pathologically confirmed renal parenchymal tumors, including 77, 46, 34, and 42 patients with clear cell renal cell carcinoma (ccRCC), chromophobe renal cell carcinoma (chRCC), angiomyolipoma (AML), and papillary renal cell carcinoma (pRCC), respectively. All enrolled patients underwent kidney MR scans with the field strength of 1.5 Tesla (T) or 3.0 T before surgery. We selected T2-weighted fat saturation sequence images of all patients and built a deep learning model to determine the type of renal tumors. Receiver operating characteristic (ROC) curve was depicted to estimate the performance of the CNN model; the accuracy, precision, sensitivity, specificity, F1-score, and area under the curve (AUC) were calculated. One-way analysis of variance and χ2 tests of independent samples were used to analyze the variables. RESULTS: The experimental results demonstrated that the model had a 60.4% overall accuracy, a 61.7% average accuracy, and a macro-average AUC of 0.82. The AUCs for ccRCC, chRCC, AML, and pRCC were 0.94, 0.78, 0.80, and 0.76, respectively. CONCLUSION: Deep CNN model based on T2-weighted fat saturation sequence MR images was useful to classify the subtypes of renal parenchymal tumors with a relatively high diagnostic accuracy.

Spatial orthogonal attention generative adversarial network for MRI reconstruction.

PURPOSE: Recent studies have witnessed that self-attention modules can better solve the vision understanding problems by capturing long-range dependencies. However, there are very few works designing a lightweight self-attention module to improve the quality of MRI reconstruction. Furthermore, it can be observed that several important self-attention modules (e.g., the non-local block) cause high computational complexity and need a huge number of GPU memory when the size of the input feature is large. The purpose of this study is to design a lightweight yet effective spatial orthogonal attention module (SOAM) to capture long-range dependencies, and develop a novel spatial orthogonal attention generative adversarial network, termed as SOGAN, to achieve more accurate MRI reconstruction. METHODS: We first develop a lightweight SOAM, which can generate two small attention maps to effectively aggregate the long-range contextual information in vertical and horizontal directions, respectively. Then, we embed the proposed SOAMs into the concatenated convolutional autoencoders to form the generator of the proposed SOGAN. RESULTS: The experimental results demonstrate that the proposed SOAMs improve the quality of the reconstructed MR images effectively by capturing long-range dependencies. Besides, compared with state-of-the-art deep learning-based CS-MRI methods, the proposed SOGAN reconstructs MR images more accurately, but with fewer model parameters. CONCLUSIONS: The proposed SOAM is a lightweight yet effective self-attention module to capture long-range dependencies, thus, can improve the quality of MRI reconstruction to a large extent. Besides, with the help of SOAMs, the proposed SOGAN outperforms the state-of-the-art deep learning-based CS-MRI methods.

TCOF1 pathogenic variants identified by Whole-exome sequencing in Chinese Treacher Collins syndrome families and hearing rehabilitation effect.

BACKGROUND: Treacher Collins syndrome (TCS, OMIM 154500) is an autosomal disorder of craniofacial development with an incidence rate of 1/50,000 live births. Although TCOF1, POLR1D, and POLR1C, have been identified as the pathogenic genes for about 90% TCS patients, the pathogenic variants of about 8-11% cases remain unknown. The object of this study is to describe the molecular basis of 14 clinically diagnosed TCS patients from four families using Whole-exome sequencing (WES) followed by Sanger sequencing confirmation, and to analyze the effect of bone conduction hearing rehabilitation in TCS patients with bilateral conductive hearing loss. RESULTS: Four previously unreported heterozygous pathogenic variants (c.3047-2A > G, c.2478 + 5G > A, c.489delC, c.648delC) were identified in the TCOF1 gene, one in each of the four families. Sanger sequencing in family members confirmed co-segregation of the identified TCOF1 variants with the phenotype. The mean pure-tone threshold improvements measured 3 months after hearing intervention were 28.8 dB for soft-band BAHA, 36.6 ± 2.0 dB for Ponto implantation, and 27.5 dB SPL for Bonebridge implantation. The mean speech discrimination improvements measured 3 months after hearing intervention in a sound field with a presentation level of 65 dB SPL were 44%, 51.25 ± 5.06, and 58%, respectively. All six patients undergoing hearing rehabilitation in this study got a satisfied hearing improvement. CONCLUSIONS: WES combined with Sanger sequencing enables the molecular diagnosis of TCS and may detect other unknown causative genes. Bone conduction hearing rehabilitation may be an optimal option for TCS patients with bilateral conductive hearing loss.

A two-step low rank matrices approach for constrained MR image reconstruction.

Low-rank structure is a powerful priori characteristic that is exploited in constrained magnetic resonance imaging (MRI). In this paper, we build two low rank matrices TV and TH from weighted k-space data according to the duality between the sparsity in the difference image and the low-rankness of a reciprocal spectral domain. Then, we propose a two-step constrained MR image reconstruction method. First, the vertical and horizontal difference images are recovered via enforcing low-rankness of matrices TV and TH. Then, the image is reconstructed via the least squares method. In the first step, the nuclear norm of a matrix is replaced by the minimum Frobenius norm of two factorization matrices and the alternating direction method of multipliers (ADMM) algorithm is applied to recover the difference images. This singular value decomposition (SVD) free method leads to fast reconstruction. The experimental results demonstrate that the proposed method outperforms other low rank based methods.

Detail-enhanced multimodality medical image fusion based on gradient minimization smoothing filter and shearing filter.

In this paper, a detail-enhanced multimodality medical image fusion algorithm is proposed by using proposed multi-scale joint decomposition framework (MJDF) and shearing filter (SF). The MJDF constructed with gradient minimization smoothing filter (GMSF) and Gaussian low-pass filter (GLF) is used to decompose source images into low-pass layers, edge layers, and detail layers at multiple scales. In order to highlight the detail information in the fused image, the edge layer and the detail layer in each scale are weighted combined into a detail-enhanced layer. As directional filter is effective in capturing salient information, so SF is applied to the detail-enhanced layer to extract geometrical features and obtain directional coefficients. Visual saliency map-based fusion rule is designed for fusing low-pass layers, and the sum of standard deviation is used as activity level measurement for directional coefficients fusion. The final fusion result is obtained by synthesizing the fused low-pass layers and directional coefficients. Experimental results show that the proposed method with shift-invariance, directional selectivity, and detail-enhanced property is efficient in preserving and enhancing detail information of multimodality medical images. Graphical abstract The detailed implementation of the proposed medical image fusion algorithm.

Image Reconstruction Using Analysis Model Prior.

The analysis model has been previously exploited as an alternative to the classical sparse synthesis model for designing image reconstruction methods. Applying a suitable analysis operator on the image of interest yields a cosparse outcome which enables us to reconstruct the image from undersampled data. In this work, we introduce additional prior in the analysis context and theoretically study the uniqueness issues in terms of analysis operators in general position and the specific 2D finite difference operator. We establish bounds on the minimum measurement numbers which are lower than those in cases without using analysis model prior. Based on the idea of iterative cosupport detection (ICD), we develop a novel image reconstruction model and an effective algorithm, achieving significantly better reconstruction performance. Simulation results on synthetic and practical magnetic resonance (MR) images are also shown to illustrate our theoretical claims.

Targeted Genes Sequencing Identified a Novel 15 bp Deletion on GJA8 in a Chinese Family with Autosomal Dominant Congenital Cataracts.

BACKGROUND: Congenital cataract (CC) is the leading cause of visual impairment or blindness in children worldwide. Because of highly genetic and clinical heterogeneity, a molecular diagnosis of the lens disease remains a challenge. METHODS: In this study, we tested a three-generation Chinese family with autosomal dominant CCs by targeted sequencing of 45 CC genes on next generation sequencing and evaluated the pathogenicity of the detected mutation by protein structure, pedigree validation, and molecular dynamics (MD) simulation. RESULTS: A novel 15 bp deletion on GJA8 (c.426_440delGCTGGAGGGGACCCT or p. 143_147delLEGTL) was detected in the family. The deletion, concerned with an in-frame deletion of 5 amino acid residues in a highly evolutionarily conserved region within the cytoplasmic loop domain of the gap junction channel protein connexin 50 (Cx50), was in full cosegregation with the cataract phenotypes in the family but not found in 1100 control exomes. MD simulation revealed that the introduction of the deletion destabilized the Cx50 gap junction channel, indicating the deletion as a dominant-negative mutation. CONCLUSIONS: The above results support the pathogenic role of the 15 bp deletion on GJA8 in the Chinese family and demonstrate targeted genes sequencing as a resolution to molecular diagnosis of CCs.

Comprehensive carrier genetic test using next-generation deoxyribonucleic acid sequencing in infertile couples wishing to conceive through assisted reproductive technology.

OBJECTIVE: To develop an expanded pan-ethnic preconception carrier genetic screening test for use in assisted reproductive technology (ART) patients and donors. DESIGN: Retrospective analysis of results obtained from 2,570 analyses. SETTING: Reproductive genetic laboratory. PATIENT(S): The 2,570 samples comprised 1,170 individuals from the gamete donor programs; 1,124 individuals corresponding to the partner of the patient receiving the donated gamete; and 276 individuals from 138 couples seeking ART using their own gametes. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Next-generation sequencing of 549 recessive and X-linked genes involved in severe childhood phenotypes reinforced with five complementary tests covering high prevalent mutations not detected by next-generation sequencing. RESULT(S): Preclinical validation included 48 DNA samples carrying known mutations for 27 genes, resulting in a sensitivity of 99%. In the clinical dataset, 2,161 samples (84%) tested positive, with an average carrier burden of 2.3 per sample. Five percent of the couples using their own gametes were found to have pathogenic variants conferring high risk for six different diseases. These high-risk couples and patients received genetic counseling and recommendations for preimplantation genetic diagnosis. For patients receiving gamete donation, we applied a genetic testing and blinded matching system to avoid high-risk combinations regardless of their carrier burden. For female donors, 1.94% were positive for X-linked conditions; they received genetic counselling and were discarded. CONCLUSION(S): We have developed a comprehensive carrier genetic screening test that, combined with our matching system and genetic counseling, constitutes a powerful tool to avoid more than 600 mendelian diseases in the offspring of patients undergoing ART.

MR image reconstruction with block sparsity and iterative support detection.

This work aims to develop a novel magnetic resonance (MR) image reconstruction approach motivated by the recently proposed sampling framework with union-of-subspaces model (SUoS). Based on SUoS, we propose a mathematical formalism that effectively integrates a block sparsity constraint and support information which is estimated in an iterative fashion. The resulting optimization problem consists of a data fidelity term and a support detection based block sparsity (SDBS) promoting term penalizing entries within the complement of the estimated support. We provide optional strategies for block assignment, and we also derive unique and robust recovery conditions in terms of the structured restricted isometric property (RIP), namely the block-RIP. The block-RIP constant we derive is lower than that of the previous structured sparse method, which leads to a reduction of the measurements. Simulation results for reconstructing individual and multiple T1/T2-weighted images demonstrate the consistency with our theoretical claims, and show considerable improvement in comparison with methods using only block sparsity or support information.

Compressed sensing MR image reconstruction exploiting TGV and wavelet sparsity.

Compressed sensing (CS) based methods make it possible to reconstruct magnetic resonance (MR) images from undersampled measurements, which is known as CS-MRI. The reference-driven CS-MRI reconstruction schemes can further decrease the sampling ratio by exploiting the sparsity of the difference image between the target and the reference MR images in pixel domain. Unfortunately existing methods do not work well given that contrast changes are incorrectly estimated or motion compensation is inaccurate. In this paper, we propose to reconstruct MR images by utilizing the sparsity of the difference image between the target and the motion-compensated reference images in wavelet transform and gradient domains. The idea is attractive because it requires neither the estimation of the contrast changes nor multiple times motion compensations. In addition, we apply total generalized variation (TGV) regularization to eliminate the staircasing artifacts caused by conventional total variation (TV). Fast composite splitting algorithm (FCSA) is used to solve the proposed reconstruction problem in order to improve computational efficiency. Experimental results demonstrate that the proposed method can not only reduce the computational cost but also decrease sampling ratio or improve the reconstruction quality alternatively.

Characterization of two long-chain fatty acid CoA ligases in the Gram-positive bacterium Geobacillus thermodenitrificans NG80-2.

The functions of two long-chain fatty acid CoA ligase genes (facl) in crude oil-degrading Geobacillus thermodenitrificans NG80-2 were characterized. Facl1 and Facl2 encoded by GTNG_0892 and GTNG_1447 were expressed in Escherichia coli and purified as His-tagged fusion proteins. Both enzymes utilized a broad range of fatty acids ranging from acetic acid (C(2)) to melissic acid (C(30)). The most preferred substrates were capric acid (C(10)) for Facl1 and palmitic acid (C(16)) for Facl2, respectively. Both enzymes had an optimal temperature of 60°C, an optimal pH of 7.5, and required ATP as a cofactor. Thermostability of the enzymes and effects of metal ions, EDTA, SDS and Triton X-100 on the enzyme activity were also investigated. When NG80-2 was cultured with crude oil rather than sucrose as the sole carbon source, upregulation of facl1 and facl2 mRNA was observed by real time RT-PCR. This is the first time that the activity of fatty acid CoA ligases toward long-chain fatty acids up to at least C(30) has been demonstrated in bacteria.

Compressed sensing MR image reconstruction using a motion-compensated reference.

Compressed sensing (CS)-based methods have been proposed for image reconstruction from undersampled magnetic resonance data. Recently, CS-based schemes using reference images have also been proposed to further reduce the sampling requirement. In this study, we propose a new reference-constrained CS reconstruction method that accounts for the misalignment between the reference and the target image to be reconstructed. The proposed method uses a new image model that represents the target image as a linear combination of a motion-dependent reference image and a sparse difference image. We then use an efficient iterative algorithm to jointly estimate the motion parameters and the difference image from sparsely sampled data. Simulation results from a numerical phantom data set and an in vivo data set show that the proposed method can accurately compensate the motion effects between the reference and the target images and improve reconstruction quality. The proposed method should prove useful for several applications such as interventional imaging, longitudinal imaging studies and dynamic contrast-enhanced imaging.

Engineering of LadA for enhanced hexadecane oxidation using random- and site-directed mutagenesis.

LadA, a monooxygenase catalyzing the oxidation of n-alkanes to 1-alkanols, is the key enzyme for the degradation of long-chain alkanes (C(15)-C(36)) in Geobacillus thermodenitrificans NG80-2. In this study, random- and site-directed mutagenesis were performed to enhance the activity of the enzyme. By screening 7,500 clones from random-mutant libraries for enhanced hexadecane hydroxylation activity, three mutants were obtained: A102D, L320V, and F146C/N376I. By performing saturation site-directed mutagenesis at the 102, 320, 146, and 376 sites, six more mutants (A102E, L320A, F146Q/N376I, F146E/N376I, F146R/N376I, and F146N/N376I) were generated. Kinetic studies showed that the hydroxylation activity of purified LadA mutants on hexadecane was 2-3.4-fold higher than that of the wild-type enzyme, with the activity of F146N/N376I being the highest. Effects of the mutations on optimum temperature, pH, and heat stability of LadA were also investigated. A complementary study showed that Pseudomonas fluorescens KOB2Δ1 strains expressing the LadA mutants grew more rapidly with hexadecane than the strain expressing wild-type LadA, confirming the enhanced activity of LadA mutants in vivo. Structural changes resulting from the mutations were analyzed and the correlation between structural changes and enzyme activity was discussed. The mutants generated in this study are potentially useful for the treatment of environmental oil pollution and in other bioconversion processes.