Automatic detection of Covid-19 from chest X-ray and lung computed tomography images using deep neural networks and transfer learning.

The world has been undergoing the most ever unprecedented circumstances caused by the coronavirus pandemic, which is having a devastating global effect in different aspects of life. Since there are not effective antiviral treatments for Covid-19 yet, it is crucial to early detect and monitor the progression of the disease, thereby helping to reduce mortality. While different measures are being used to combat the virus, medical imaging techniques have been examined to support doctors in diagnosing the disease. In this paper, we present a practical solution for the detection of Covid-19 from chest X-ray (CXR) and lung computed tomography (LCT) images, exploiting cutting-edge Machine Learning techniques. As the main classification engine, we make use of EfficientNet and MixNet, two recently developed families of deep neural networks. Furthermore, to make the training more effective and efficient, we apply three transfer learning algorithms. The ultimate aim is to build a reliable expert system to detect Covid-19 from different sources of images, making it be a multi-purpose AI diagnosing system. We validated our proposed approach using four real-world datasets. The first two are CXR datasets consist of 15,000 and 17,905 images, respectively. The other two are LCT datasets with 2,482 and 411,528 images, respectively. The five-fold cross-validation methodology was used to evaluate the approach, where the dataset is split into five parts, and accordingly the evaluation is conducted in five rounds. By each evaluation, four parts are combined to form the training data, and the remaining one is used for testing. We obtained an encouraging prediction performance for all the considered datasets. In all the configurations, the obtained accuracy is always larger than 95.0%. Compared to various existing studies, our approach yields a substantial performance gain. Moreover, such an improvement is statistically significant.

Identification of rare high-risk copy number variants affecting the dopamine transporter gene in mental disorders.

BACKGROUND: The dopamine transporter, also known as solute carrier 6A3 (SLC6A3), plays an important role in synaptic transmission by regulating the reuptake of dopamine in the synapses. In line with this, variations in the gene encoding this transporter have been linked to both schizophrenia and affective disorders. Recently, copy number variants (CNVs) in SLC6A3 have been identified in healthy subjects but so far, the implication of CNVs affecting this gene in psychiatric diseases has not been addressed. AIMS: In the present study, we aimed to investigate whether CNVs affecting SLC6A3 represent rare high-risk variants of psychiatric disorders. METHODS: We performed a systematic screening for CNVs affecting SLC6A3 in 761 healthy controls, 672 schizophrenia patients, and 194 patients with bipolar disorder in addition to 253 family members from six large pedigrees affected by mental disorders using single nucleotide polymorphism arrays and subsequent verification by real-time polymerase chain reaction. RESULTS: We identified two duplications and one deletion affecting SLC6A3 in the patients, while no such CNVs were identified in any of the controls. The identified CNVs were of different sizes and two affected several genes in addition to SLC6A3. CONCLUSION: Our findings suggest that rare high-risk CNVs affecting the gene encoding the dopamine transporter contribute to the pathogenesis of schizophrenia and affective disorders.

Two rare deletions upstream of the NRXN1 gene (2p16.3) affecting the non-coding mRNA AK127244 segregate with diverse psychopathological phenotypes in a family.

CNVs spanning the 2p16.3 (NRXN1) and the 15q11.2 gene rich region have been associated with severe neuropsychiatric disorders including schizophrenia. Recently, studies have also revealed that CNVs in non-coding regions play an essential role in genomic variability in addition to disease susceptibility. In this study, we describe a family affected by a wide range of psychiatric disorders including early onset schizophrenia, schizophreniform disorder, and affective disorders. Microarray analysis identified two rare deletions immediately upstream of the NRXN1 gene affecting the non-coding mRNA AK127244 in addition to the pathogenic 15q11.2 deletion in distinct family members. The two deletions upstream of the NRXN1 gene were found to segregate with psychiatric disorders in the family and further similar deletions have been observed in patients diagnosed with autism spectrum disorder. Thus, we suggest that non-coding regions upstream of the NRXN1 gene affecting AK127244 might (as NRXN1) contain susceptibility regions for a wide spectrum of neuropsychiatric disorders.