Developing global image feature analysis models to predict cancer risk and prognosis

In order to develop precision or personalized medicine,identifying new quantitative imaging markers and building machine learning models to predict cancer risk and prognosis has been attracting broad research interest recently.Most of these research approaches use the similar concepts of the conventional computer-aided detection schemes of medical images,which include steps in detecting and segmenting suspicious regions or tumors,followed by training machine learning models based on the fusion of multiple image features computed from the segmented regions or tumors.However,due to the heterogeneity and boundary fuzziness of the suspicious regions or tumors,segmenting subtle regions is often difficult and unreliable.Additionally,ignoring global and/or background parenchymal tissue characteristics may also be a limitation of the conventional approaches.In our recent studies,we investigated the feasibility of developing new computer-aided schemes implemented with the machine learning models that are trained by global image features to predict cancer risk and prognosis.We trained and tested several models using images obtained from full-field digital mammography,magnetic resonance imaging,and computed tomography of breast,lung,and ovarian cancers.Study results showed that many of these new models yielded higher performance than other approaches used in current clinical practice.Furthermore,the computed global image features also contain complementary information from the features computed from the segmented regions or tumors in predicting cancer prognosis.Therefore,the global image features can be used alone to develop new case-based prediction models or can be added to current tumor-based models to increase their discriminatory power.

Background removal from global auroral images:Data-driven dayglow modeling

Global images of auroras obtained by cameras on spacecraft are a key tool for studying the near-Earth environment.However,the cameras are sensitive not only to auroral emissions produced by precipitating particles,but also to dayglow emissions produced by photoelectrons induced by sunlight.Nightglow emissions and scattered sunlight can contribute to the background signal.To fully utilize such images in space science,background contamination must be removed to isolate the auroral signal.Here we outline a data-driven approach to modeling the background intensity in multiple images by formulating linear inverse problems based on B-splines and spherical harmonics.The approach is robust,flexible,and iteratively deselects outliers,such as auroral emissions.The final model is smooth across the terminator and accounts for slow temporal variations and large-scale asymmetries in the dayglow.We demonstrate the model by using the three far ultraviolet cameras on the Imager for Magnetopause-to-Aurora Global Exploration(IMAGE)mission.The method can be applied to historical missions and is relevant for upcoming missions,such as the Solar wind Magnetosphere Ionosphere Link Explorer(SMILE)mission.

Global point tracking based panoramic image stabilization system

A novel image stabilization system is presented,which consists of a global feature point tracking based motion estimation,a Kalman filtering based motion smoothing and an image mosaic based panoramic compensation.The global motion is estimated using feature point matching and iteration with the least-square method.Then,the Kalman filter is applied to smooth the original motion vectors to effectively alleviate unwanted camera vibrations and follow the intentional camera scan.Lastly,the loss information of image boundary due to the motion compensation is reconstructed with image mosaic to improve the visual quality.The experimental results show that this system can smooth unwanted translation or rotation of the video sequences and realize a panoramic stabilization at real-time speed.

Effect of interplanetary shock on an ongoing substorm:Simultaneous satellite-ground auroral observations

Substorm processes have been studied in detail,and it is well known that interplanetary(IP)shock encountering the terrestrial magnetosphere causes global responses.However,how IP shock compression to the magnetosphere affects the development of an ongoing substorm remains uninvestigated.Herein,the simultaneous satellite and ground-based auroral evolutions associated with an IP shock impact on the magnetopause during an ongoing substorm on May 7th,2005,were examined.The IMAGE satellite over the Southern Hemisphere captured the global development substorm,which was initiated at 17:38:47 UT.The poleward branch of the nightside auroral oval was fortuitously monitored by an all-sky camera at the Zhongshan Station(-74.5°magnetic latitude,ZHO)in Antarctica.The satellite imager observed continuous brightening and broadening of the nightside auroral oval after the IP shock arrival.The simultaneous ground-based optical aurora measurement displayed the intensification and expansion of a preexisting auroral surge poleward of the aurora oval.The geomagnetic field variations and the instantly increased PC indices indicated an elevated merging rate and enhanced the convection-related DP-2 currents.Therefore,this IP shock transient impact did not significantly change the ongoing development of the substorm,although it meets the magnetospheric precondition hypothesis.

A Thorough Evaluation of the Passive Microwave Radiometer Measurements onboard Three Fengyun-3 Satellites

Microwave Radiometer Imager(MWRI) is a key payload of China’s second generation polar meteorological satellite, i.e., Fengyun-3 series(FY-3). Up to now, 5 satellites including FY-3A(2008), FY-3B(2010), FY-3C(2013), FY-3D(2018), and FY-3E(2021) have been launched successfully to provide multiwavelength, all-weather, and global data for decades. Much progress has been made on the calibration of MWRI and a recalibrated MWRI brightness temperature(BT) product(V2) was recently released. This study thoroughly evaluates the accuracy of this new product from FY-3B, 3C, and 3D by using the simultaneous collocated Global Precipitation Measurement(GPM)Microwave Imager(GMI) measurements as a reference. The results show that the mean biases(MBEs) of the BT between MWRI and GMI are generally less than 0.5 K and the root mean squares(RMSs) between them are less than1.5 K. The previous notable ascending and descending difference of the MWRI has disappeared. This indicates that the new MWRI recalibration procedure is very effective in removing potential errors associated with the emission of the hot-load reflector. Analysis of the dependence of MBE on the latitude and earth scene temperature shows that MBE decreases with decreasing latitude over ocean. Furthermore, MBE over ocean decreases linearly with increasing scene temperature for almost all channels, whereas this does not occur over land. A linear regression fitting is then used to modify MWRI, which can reduce the MBE over ocean to be within 0.2 K. The standard deviation of error of GMI, FY-3B, and FY-3D MWRI BT data derived by using the three-cornered hat method(TCH) shows that GMI has the best overall performance over ocean except at 10.65 GHz where its standard deviation of error is slightly larger than that of FY-3D. Over land, the standard deviation of error of FY-3D is the lowest at almost all channels except at 89V. MWRI onboard FY-3 series satellites would serve as an important passive microwave radiometer member of the constellation to monitor key surface and atmospheric properties.