An optimal brain tumor segmentation algorithm for clinical MRI dataset with low resolution and non-contiguous slices.

BACKGROUND: Segmenting brain tumor and its constituent regions from magnetic resonance images (MRI) is important for planning diagnosis and treatment. In clinical routine often an experienced radiologist delineates the tumor regions using multimodal MRI. But this manual segmentation is prone to poor reproducibility and is time consuming. Also, routine clinical scans are usually of low resolution. To overcome these limitations an automated and precise segmentation algorithm based on computer vision is needed. METHODS: We investigated the performance of three widely used segmentation methods namely region growing, fuzzy C means and deep neural networks (deepmedic). We evaluated these algorithms on the BRATS 2018 dataset by choosing randomly 48 patients data (high grade, n = 24 and low grade, n = 24) and on our routine clinical MRI brain tumor dataset (high grade, n = 15 and low grade, n = 28). We measured their performance using dice similarity coefficient, Hausdorff distance and volume measures. RESULTS: Region growing method performed very poorly when compared to fuzzy C means (fcm) and deepmedic network. Dice similarity coefficient scores for FCM and deepmedic algorithms were close to each other for BRATS and clinical dataset. The accuracy was below 70% for both these methods in general. CONCLUSION: Even though the deepmedic network showed very high accuracy in BRATS challenge for brain tumor segmentation, it has to be custom trained for the low resolution routine clinical scans. It also requires large training data to be used as a stand-alone algorithm for clinical applications. Nevertheless deepmedic may be a better algorithm for brain tumor segmentation when compared to region growing or FCM.

Plasmonic Ag nanoparticles arbitrated enhanced photodetection in p-NiO/n-rGO heterojunction for future self-powered UV photodetectors.

We report on the low cost and low temperature chemical synthesis of p-type nickel oxide (NiO) and n-type reduced graphene oxide (rGO) and their integration onto ITO/glass substrate to form p-NiO/n-rGO heterojunction for possible self-powered ultraviolet (UV) photodetector applications. Different spectroscopies and microscopes were employed to study their microstructural and surface properties. Whereas, the electrical characterizations have been performed on the devices to ascertain the responsivity, detectivity, external quantum efficiency and temporal responses under dark and UV illumination. It is noteworthy that rGO has not only been used as an n-type semiconductor, but also acted as an electron transport layer, which satisfactorily separates out the electrons from the generated carrier pairs, leading to enhanced photoresponse. Furthermore, efforts were also consecrated to synthesize Ag nanoparticles (NPs) of ∼5 nm radius. The integration of Ag NPs on the conventional NiO/rGO heterojunction facilitates an improved UV light absorption property. It was understood that the performance improvement was owed to the local surface plasmon resonance of Ag NPs within the active layer of NiO. Surprisingly, both the devices (with and without Ag NPs) exhibit photovoltaic behavior which shows its potential for self-powered device application. When the Ag NPs embedded device is concerned, it showed better on/off ratio (6.3 × 103), high responsivity (72 mAW-1), large detectivity (3.95 × 1012 Jones), and high efficiency (24.46%) as compared to the conventional NiO/rGO heterojunction one (without Ag NPs). The variation in the photoresponse and improved charge transport was explained through a band-diagram, which also showcases a comprehensive understanding on the operational principle of the fabricated self-powered devices. Thus, this self-powered photodetector driven by built in electric field is operated independently and can be attached with any other electronic gadgets for internet of things applications.