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

Radiology Radiology

Development of a volumetric pancreas segmentation CT dataset for AI applications through trained technologists: a study during the COVID 19 containment phase.

In Abdominal radiology (New York)

PURPOSE : To evaluate the performance of trained technologists vis-à-vis radiologists for volumetric pancreas segmentation and to assess the impact of supplementary training on their performance.

METHODS : In this IRB-approved study, 22 technologists were trained in pancreas segmentation on portal venous phase CT through radiologist-led interactive videoconferencing sessions based on an image-rich curriculum. Technologists segmented pancreas in 188 CTs using freehand tools on custom image-viewing software. Subsequent supplementary training included multimedia videos focused on common errors, which were followed by second batch of 159 segmentations. Two radiologists reviewed all cases and corrected inaccurate segmentations. Technologists' segmentations were compared against radiologists' segmentations using Dice-Sorenson coefficient (DSC), Jaccard coefficient (JC), and Bland-Altman analysis.

RESULTS : Corrections were made in 71 (38%) cases from first batch [26 (37%) oversegmentations and 45 (63%) undersegmentations] and in 77 (48%) cases from second batch [12 (16%) oversegmentations and 65 (84%) undersegmentations]. DSC, JC, false positive (FP), and false negative (FN) [mean (SD)] in first versus second batches were 0.63 (0.15) versus 0.63 (0.16), 0.48 (0.15) versus 0.48 (0.15), 0.29 (0.21) versus 0.21 (0.10), and 0.36 (0.20) versus 0.43 (0.19), respectively. Differences were not significant (p > 0.05). However, range of mean pancreatic volume difference reduced in the second batch [- 2.74 cc (min - 92.96 cc, max 87.47 cc) versus - 23.57 cc (min - 77.32, max 30.19)].

CONCLUSION : Trained technologists could perform volumetric pancreas segmentation with reasonable accuracy despite its complexity. Supplementary training further reduced range of volume difference in segmentations. Investment into training technologists could augment and accelerate development of body imaging datasets for AI applications.

Suman Garima, Panda Ananya, Korfiatis Panagiotis, Edwards Marie E, Garg Sushil, Blezek Daniel J, Chari Suresh T, Goenka Ajit H


Artificial intelligence, COVID-19, Data curation, Deep learning

General General

Bcr-Abl Allosteric Inhibitors: Where We Are and Where We Are Going to.

In Molecules (Basel, Switzerland)

The fusion oncoprotein Bcr-Abl is an aberrant tyrosine kinase responsible for chronic myeloid leukemia and acute lymphoblastic leukemia. The auto-inhibition regulatory module observed in the progenitor kinase c-Abl is lost in the aberrant Bcr-Abl, because of the lack of the N-myristoylated cap able to bind the myristoyl binding pocket also conserved in the Bcr-Abl kinase domain. A way to overcome the occurrence of resistance phenomena frequently observed for Bcr-Abl orthosteric drugs is the rational design of allosteric ligands approaching the so-called myristoyl binding pocket. The discovery of these allosteric inhibitors although very difficult and extremely challenging, represents a valuable option to minimize drug resistance, mostly due to the occurrence of mutations more frequently affecting orthosteric pockets, and to enhance target selectivity with lower off-target effects. In this perspective, we will elucidate at a molecular level the structural bases behind the Bcr-Abl allosteric control and will show how artificial intelligence can be effective to drive the automated de novo design towards off-patent regions of the chemical space.

Carofiglio Francesca, Trisciuzzi Daniela, Gambacorta Nicola, Leonetti Francesco, Stefanachi Angela, Nicolotti Orazio


Bcr-Abl, allosteric inhibitors, artificial intelligence, chronic myeloid leukemia, de novo design

General General

Investigating the Capabilities of Information Technologies to support Policymaking in COVID-19 Crisis Management; A Systematic Review and Expert opinions.

In European journal of clinical investigation

BACKGROUND : Today, numerous countries are fighting to protect themselves against the Covid-19 crisis, while the policymakers are confounded and empty-handed in dealing with this chaotic circumstance. The infection and its impacts have made it difficult to make optimal and suitable decisions. New information technologies play significant roles in such critical situations to address and relieve stress during the coronavirus crisis. This article endeavors to recognize the challenges policymakers have typically experienced during pandemic diseases, including Covid-19, and, accordingly, new information technology capabilities to encounter with them.

MATERIAL AND METHODS : The current study utilizes the synthesis of findings of experts' opinions within the systematic review process as the research method to recognize the best available evidence drawn from text and opinion to offer practical guidance for policymakers.

RESULTS : The results illustrate that the challenges fall into two categories including; encountering the disease and reducing the results of the disease. Furthermore, Internet of Things, cloud computing, machine learning, and social networking play the most significant roles to address these challenges.

Lagzian Mohammad, Dadkhah Mehdi, Mehraeen AmirReza


Covid-19, Crisis management policies, Informational Technology (IT) capabilities, Pandemic management

General General

Time-series Imputation and Prediction with Bi-Directional Generative Adversarial Networks

ArXiv Preprint

Multivariate time-series data are used in many classification and regression predictive tasks, and recurrent models have been widely used for such tasks. Most common recurrent models assume that time-series data elements are of equal length and the ordered observations are recorded at regular intervals. However, real-world time-series data have neither a similar length nor a same number of observations. They also have missing entries, which hinders the performance of predictive tasks. In this paper, we approach these issues by presenting a model for the combined task of imputing and predicting values for the irregularly observed and varying length time-series data with missing entries. Our proposed model (Bi-GAN) uses a bidirectional recurrent network in a generative adversarial setting. The generator is a bidirectional recurrent network that receives actual incomplete data and imputes the missing values. The discriminator attempts to discriminate between the actual and the imputed values in the output of the generator. Our model learns how to impute missing elements in-between (imputation) or outside of the input time steps (prediction), hence working as an effective any-time prediction tool for time-series data. Our method has three advantages to the state-of-the-art methods in the field: (a) single model can be used for both imputation and prediction tasks; (b) it can perform prediction task for time-series of varying length with missing data; (c) it does not require to know the observation and prediction time window during training which provides a flexible length of prediction window for both long-term and short-term predictions. We evaluate our model on two public datasets and on another large real-world electronic health records dataset to impute and predict body mass index (BMI) values in children and show its superior performance in both settings.

Mehak Gupta, Rahmatollah Beheshti


Radiology Radiology

Deep learning method for segmentation of rotator cuff muscles on MR images.

In Skeletal radiology

OBJECTIVE : To develop and validate a deep convolutional neural network (CNN) method capable of (1) selecting a specific shoulder sagittal MR image (Y-view) and (2) automatically segmenting rotator cuff (RC) muscles on a Y-view. We hypothesized a CNN approach can accurately perform both tasks compared with manual reference standards.

MATERIAL AND METHODS : We created 2 models: model A for Y-view selection and model B for muscle segmentation. For model A, we manually selected shoulder sagittal T1 Y-views from 258 cases as ground truth to train a classification CNN (Keras/Tensorflow, Inception v3, 16 batch, 100 epochs, dropout 0.2, learning rate 0.001, RMSprop). A top-3 success rate evaluated model A on 100 internal and 50 external test cases. For model B, we manually segmented subscapularis, supraspinatus, and infraspinatus/teres minor on 1048 sagittal T1 Y-views. After histogram equalization and data augmentation, the model was trained from scratch (U-Net, 8 batch, 50 epochs, dropout 0.25, learning rate 0.0001, softmax). Dice (F1) score determined segmentation accuracy on 105 internal and 50 external test images.

RESULTS : Model A showed top-3 accuracy > 98% to select an appropriate Y-view. Model B produced accurate RC muscle segmentations with mean Dice scores > 0.93. Individual muscle Dice scores on internal/external datasets were as follows: subscapularis 0.96/0.93, supraspinatus 0.97/0.96, and infraspinatus/teres minor 0.97/0.95.

CONCLUSIONS : Our results show overall accurate Y-view selection and automated RC muscle segmentation using a combination of deep CNN algorithms.

Medina Giovanna, Buckless Colleen G, Thomasson Eamon, Oh Luke S, Torriani Martin


Artificial intelligence, Atrophy, MRI, Muscles, Rotator cuff, Segmentation, Shoulder

Pathology Pathology

Research advances in forensic diatom testing.

In Forensic sciences research

In forensic practice, it is difficult to determine whether a dead body in the water resulted from drowning or from disposal after death. Diatom testing is currently an important supporting technique for the determination of death by drowning and of drowning sites, even though it is a time-consuming and laborious task. This article reviews the development of diatom testing over the decades and discusses a new method for the potential application of deep learning in diatom testing.

Zhou Yuanyuan, Cao Yongjie, Huang Jiao, Deng Kaifei, Ma Kaijun, Zhang Tianye, Chen Liqin, Zhang Ji, Huang Ping


Forensic sciences, deep learning, diatom, drowning, forensic pathology