Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Generative Imputation and Stochastic Prediction.

In IEEE transactions on pattern analysis and machine intelligence ; h5-index 127.0

In many machine learning applications, we are faced with incomplete datasets. In the literature, missing data imputation techniques have been mostly concerned with filling missing values. However, the existence of missing values is synonymous with uncertainties not only over the distribution of missing values but also over target class assignments that require careful consideration. In this paper, we propose a simple and effective method for imputing missing features and estimating the distribution of target assignments given incomplete data. In order to make imputations, we train a simple and effective generator network to generate imputations that a discriminator network is tasked to distinguish. Following this, a predictor network is trained using the imputed samples from the generator network to capture the classification uncertainties and make predictions accordingly. The proposed method is evaluated on CIFAR-10 and MNIST image datasets as well as five real-world tabular classification datasets, under different missingness rates and structures. Our experimental results show the effectiveness of the proposed method in generating imputations as well as providing estimates for the class uncertainties in a classification task when faced with missing values.

Kachuee Mohammad, Karkkainen Kimmo, Goldstein Orpaz, Darabi Sajad, Sarrafzadeh Majid

2020-Sep-07

General General

Glial Modulation of Electrical Rhythms in a Neuroglial Network Model of Epilepsy.

In IEEE transactions on bio-medical engineering

OBJECTIVE : An important EEG-based biomarker for epilepsy is the phase-amplitude cross-frequency coupling (PAC) of electrical rhythms; however, the underlying pathways of these pathologic markers are not always clear. Since glial cells have been shown to play an active role in neuroglial networks, it is likely that some of these PAC markers are modulated via glial effects.

METHODS : We developed a 4-unit hybrid model of a neuroglial network, consisting of 16 sub-units, that combines a mechanistic representation of neurons with an oscillator- based Cognitive Rhythm Generator (CRG) representation of glial cells astrocytes and microglia. The model output was compared with recorded generalized tonic-clonic patient data, both in terms of PAC features, and state classification using an unsupervised hidden Markov model (HMM).

RESULTS : The neuroglial model output showed PAC features similar to those observed in epileptic seizures. These generated PAC features were able to accurately identify spontaneous epileptiform discharges (SEDs) as seizure- like states, as well as a postictal-like state following the long- duration SED, when applied to the HMM machine learning algorithm trained on patient data. The evolution profile of the maximal PAC during the SED compared well with patient data, showing similar association with the duration of the postictal state.

CONCLUSION : The hybrid neuroglial network model was able to generate PAC features similar to those observed in ictal and postictal epileptic states, which has been used for state classification and postictal state duration prediction.

SIGNIFICANCE : Since PAC biomarkers are important for epilepsy research and postictal state duration has been linked with risk of sudden unexplained death in epilepsy, this model suggests glial synaptic effects as potential targets for further analysis and treatment.

Grigorovsky Vasily, Breton Vanessa, Bardakjian Berj L

2020-Sep-07

Surgery Surgery

Letter to Editor: Predicting survival after hepatocellular carcinoma resection using deep-learning on histological slides.

In Hepatology (Baltimore, Md.)

We read with interest the article by Saillard et al [1] which developed two deep-learning models based on whole-slide digitized histological slides (WSI) to predict the survival of patients with HCC treated by surgical resection. The study showed the models have a greater discriminatory power in two independent cohorts than an optimal combination of all classical and relevant clinical, biological and pathological features. We also appreciate their effort for insights into the most relevant features used by the network for prediction.

Zhen Shihui, Cai Xiujun

2020-Sep-07

General General

Benchmarking off-the-shelf statistical shape modeling tools in clinical applications

ArXiv Preprint

Statistical shape modeling (SSM) is widely used in biology and medicine as a new generation of morphometric approaches for the quantitative analysis of anatomical shapes. Technological advancements of in vivo imaging have led to the development of open-source computational tools that automate the modeling of anatomical shapes and their population-level variability. However, little work has been done on the evaluation and validation of such tools in clinical applications that rely on morphometric quantifications (e.g., implant design and lesion screening). Here, we systematically assess the outcome of widely used, state-of-the-art SSM tools, namely ShapeWorks, Deformetrica, and SPHARM-PDM. We use both quantitative and qualitative metrics to evaluate shape models from different tools. We propose validation frameworks for anatomical landmark/measurement inference and lesion screening. We also present a lesion screening method to objectively characterize subtle abnormal shape changes with respect to learned population-level statistics of controls. Results demonstrate that SSM tools display different levels of consistencies, where ShapeWorks and Deformetrica models are more consistent compared to models from SPHARM-PDM due to the groupwise approach of estimating surface correspondences. Furthermore, ShapeWorks and Deformetrica shape models are found to capture clinically relevant population-level variability compared to SPHARM-PDM models.

Anupama Goparaju, Alexandre Bone, Nan Hu, Heath B. Henninger, Andrew E. Anderson, Stanley Durrleman, Matthijs Jacxsens, Alan Morris, Ibolya Csecs, Nassir Marrouche, Shireen Y. Elhabian

2020-09-07

Surgery Surgery

Brain Tumor Survival Prediction using Radiomics Features

ArXiv Preprint

Surgery planning in patients diagnosed with brain tumor is dependent on their survival prognosis. A poor prognosis might demand for a more aggressive treatment and therapy plan, while a favorable prognosis might enable a less risky surgery plan. Thus, accurate survival prognosis is an important step in treatment planning. Recently, deep learning approaches have been used extensively for brain tumor segmentation followed by the use of deep features for prognosis. However, radiomics-based studies have shown more promise using engineered/hand-crafted features. In this paper, we propose a three-step approach for multi-class survival prognosis. In the first stage, we extract image slices corresponding to tumor regions from multiple magnetic resonance image modalities. We then extract radiomic features from these 2D slices. Finally, we train machine learning classifiers to perform the classification. We evaluate our proposed approach on the publicly available BraTS 2019 data and achieve an accuracy of 76.5% and precision of 74.3% using the random forest classifier, which to the best of our knowledge are the highest reported results yet. Further, we identify the most important features that contribute in improving the prediction.

Sobia Yousaf, Syed Muhammad Anwar, Harish RaviPrakash, Ulas Bagci

2020-09-07

General General

A novel action recognition system for smart monitoring of elderly people using Action Pattern Image and Series CNN with transfer learning

ArXiv Preprint

Falling of elderly people who are staying alone at home leads to health risks. If they are not attended immediately even it may lead to fatal danger to their life. In this paper a novel computer vision-based system for smart monitoring of elderly people using Series Convolutional Neural Network (SCNN) with transfer learning is proposed. When CNN is trained by the frames of the videos directly, it learns from all pixels including the background pixels. Generally, the background in a video does not contribute anything in identifying the action and actually it will mislead the action classification. So, we propose a novel action recognition system and our contributions are 1) to generate more general action patterns which are not affected by illumination and background variations of the video sequences and eliminate the obligation of image augmentation in CNN training 2) to design SCNN architecture and enhance the feature extraction process to learn large amount of data, 3) to present the patterns learnt by the neurons in the layers and analyze how these neurons capture the action when the input pattern is passing through these neurons, and 4) to extend the capability of the trained SCNN for recognizing fall actions using transfer learning.

L. Aneesh Euprazia, K. K. Thyagharajan

2020-09-07