Machine learning classifies predictive kinematic features in a mouse model of neurodegeneration

AbstractMotor deficits are observed in Alzheimer’s disease (AD) prior to the appearance of cognitive symptoms. To investigate the role of amyloid proteins in gait disturbances, we characterized locomotion in APP-overexpressing transgenic J20 mice. We used three-dimensional motion capture to characterize quadrupedal locomotion on a treadmill in J20 and wild-type mice. Sixteen J20 mice and fifteen wild-type mice were studied at two ages (4- and 13-month). A random forest (RF) classification algorithm discriminated between the genotypes within each age group using a leave-one-out cross-validation. The balanced accuracy of the RF classification was 92.3 ± 5.2% and 93.3 ± 4.5% as well as False Negative Rate (FNR) of 0.0 ± 0.0% and 0.0 ± 0.0% for the 4-month and 13-month groups, respectively. Feature ranking algorithms identified kinematic features that when considered simultaneously, achieved high genotype classification accuracy. The identified features demonstrated an age-specific kinematic profile of the impact of APP-overexpression. Trunk tilt and unstable hip movement patterns were important in classifying the 4-month J20 mice, whereas patterns of shoulder and iliac crest movement were critical for classifying 13-month J20 mice. Examining multiple kinematic features of gait simultaneously could also be developed to classify motor disorders in humans.

Download Full-text

Abstract 1122‐000034: Systematic CT‐Perfusion acquisition in All Suspected Stroke Patients Increases Vascular Occlusion Detection and Thrombectomy Rates

Stroke: Vascular and Interventional Neurology ◽

10.1161/svin.01.suppl_1.000034 ◽

2021 ◽

Vol 1 (S1) ◽

Author(s):

Marta Olive‐Gadea ◽

Manuel Requena ◽

Facundo Diaz ◽

Alvaro Garcia‐Tornel ◽

Marta Rubiera ◽

...

Keyword(s):

Ischemic Stroke ◽

Acute Ischemic Stroke ◽

Clinical Symptoms ◽

Ct Perfusion ◽

False Negative ◽

False Negative Rate ◽

Vascular Occlusion ◽

Stroke Patients ◽

Imaging Protocols ◽

The Impact

Introduction : In acute ischemic stroke patients, current guidelines recommend noninvasive vascular imaging to identify intracranial vessel occlusions (VO) that may benefit from endovascular treatment (EVT). However, VO can be missed in CT angiography (CTA) readings. We aim to evaluate the impact of consistently including CT perfusion (CTP) in admission stroke imaging protocols on VO diagnosis and EVT rates. Methods : We included patients with a suspected acute ischemic stroke that underwent urgent non‐contrast CT, CTA and CTP from April to October 2020. Hypoperfusion areas defined by Tmax>6s delay (RAPID software), congruent with the clinical symptoms and a vascular territory, were considered due to a VO (CTP‐VO). Cases in which mechanical thrombectomy was performed were defined as therapeutically relevant VO (EVT‐VO). For patients that received EVT, site of VO according to digital subtraction angiography was recorded. Two experienced neuroradiologists blinded to CTP but not to clinical symptoms, retrospectively evaluated NCCT and CTA to identify intracranial VO (CTA‐VO). We analyzed CTA‐VO sensitivity and specificity at detecting CTP‐VO and EVT‐VO respecitvely. We performed a logistic regression to test the association of Tmax>6s volumes with CTA‐VO identification and indication of EVT. Results : Of the 338 patients included in the analysis, 157 (46.5%) presented a CTP‐VO, (median Tmax>6s: 73 [29‐127] ml). CTA‐VO was identified in 83 (24.5%) of the cases. Overall CTA‐VO sensitivity for the detection of CTP‐VO was 50.3% and specificity was 97.8%. Higher hypoperfusion volume was associated with an increased CTA‐VO detection, with an odds ratio of 1.03 (95% confidence interval 1.02‐1.04) (figure). DSA was indicated in 107 patients; in 4 of them no EVT was attempted due to recanalization or a too distal VO in the first angiographic run. EVT was performed in 103 patients (30.5%. Tmax>6s: 102 [63‐160] ml), representing 65.6% of all CTP‐VO. Overall CTA‐VO sensitivity for the detection of EVT‐VO was 69.9%. The CTA‐VO sensitivity for detecting patients with indication of EVT according to clinical guidelines was as follows: 91.7% for ICA occlusions and 84.4% for M1‐MCA occlusions. For all other occlusion sites that received EVT, the CTA‐VO sensitivity was 36.1%. The overall specificity was 95.3%. Among patients who received EVT, CTA‐VO was not detected in 31 cases, resulting in a false negative rate of 30.1%. False negative CTA‐VO cases had lower Tmax>6s volumes (69[46‐99.5] vs 126[84‐169.5]ml, p<0.001) and lower NIHSS (13[8.5‐16] vs 17[14‐21], p<0.001). Conclusions : Systematically including CTP perfusion in the acute stroke admission imaging protocols may increase the diagnosis of VO and rate of EVT.

Download Full-text

Three-dimensional camera anthropometry to assess risk of cephalopelvic disproportion-related obstructed labour in Ethiopia

Interface Focus ◽

10.1098/rsfs.2019.0036 ◽

2019 ◽

Vol 9 (5) ◽

pp. 20190036 ◽

Cited By ~ 1

Author(s):

Lorenzo Tolentino ◽

Mahlet Yigeremu ◽

Sisay Teklu ◽

Shehab Attia ◽

Michael Weiler ◽

...

Keyword(s):

False Negative ◽

Three Dimensional ◽

False Negative Rate ◽

Detection Rates ◽

Cephalopelvic Disproportion ◽

Obstructed Labour ◽

Tape Measure ◽

Negative Rate ◽

Positive Rate ◽

Tape Measurement

Cephalopelvic disproportion (CPD)-related obstructed labour requires delivery via Caesarean section (C/S); however, in low-resource settings around the world, facilities with C/S capabilities are often far away. This paper reports three low-cost tools to assess the risk of CPD, well before labour, to provide adequate time for referral and planning for delivery. We performed tape measurement- and three-dimensional (3D) camera-based anthropometry, using two 3D cameras (Kinect and Structure) on primigravida, gestational age ≥ 36 weeks, from Addis Ababa, Ethiopia. Novel risk scores were developed and tested to identify models with the highest predicted area under the receiver-operator characteristic curve (AUC), detection rate (true positive rate at a 5% false-positive rate, FPR) and triage rate (true negative rate at a 0% false-negative rate). For tape measure, Kinect and Structure, the detection rates were 53%, 61% and 64% (at 5% FPR), the triage rates were 30%, 56% and 63%, and the AUCs were 0.871, 0.908 and 0.918, respectively. Detection rates were 77%, 80% and 84% at the maximum J -statistic, which corresponded to FPRs of 10%, 15% and 11%, respectively, for tape measure, Kinect and Structure. Thus, tape measurement anthropometry was a very good predictor and Kinect and Structure anthropometry were excellent predictors of CPD risk.

Download Full-text

Secondary Complete Cytoreduction in Recurrent Ovarian Cancer: Benefit of Optimal Patient Selection Using Scoring System

International Journal of Gynecological Cancer ◽

10.1097/igc.0000000000000051 ◽

2014 ◽

Vol 24 (2) ◽

pp. 238-246 ◽

Cited By ~ 15

Author(s):

Enora Laas ◽

Mathieu Luyckx ◽

Marjolein De Cuypere ◽

Frederic Selle ◽

Emile Daraï ◽

...

Keyword(s):

Ovarian Cancer ◽

Patient Selection ◽

Positive Impact ◽

False Negative ◽

False Negative Rate ◽

Recurrent Ovarian Cancer ◽

Complete Resection ◽

Predictive Values ◽

Independent Population ◽

The Impact

ObjectiveComplete tumor cytoreduction seems to be beneficial for patients with recurrent epithelial ovarian cancer (REOC). The challenge is to identify patients eligible for such surgery. Several scores based on simple clinical parameters have attempted to predict resectability and help in patient selection for surgery in REOC.The aims of this study were to assess the performance of these models in an independent population and to evaluate the impact of complete resection.Materials and MethodsA total of 194 patients with REOC between January 2000 and December 2010 were included in 2 French centers. Two scores were used: the AGO DESKTOP OVAR trial score and a score from Tian et al.The performance (sensitivity, specificity, and predictive values) of these scores was evaluated in our population. Survival curves were constructed to evaluate the survival impact of surgery on recurrence.ResultsPositive predictive values for complete resection were 80.6% and 74.0% for the DESKTOP trial score and the Tian score, respectively. The false-negative rate was high for both models (65.4% and 71.4%, respectively). We found a significantly higher survival in the patients with complete resection (59.4 vs 17.9 months,P< 0.01) even after adjustment for the confounding variables (hazard ratio [HR], 2.53; 95% confidence interval, 1.01–6.3;P= 0.04).ConclusionsIn REOC, surgery seems to have a positive impact on survival, if complete surgery can be achieved. However, factors predicting complete resection are not yet clearly defined. Recurrence-free interval and initial resection seem to be the most relevant factors. Laparoscopic evaluation could help to clarify the indications for surgery.

Download Full-text

Hidden biases in germline structural variant detection

Genome Biology ◽

10.1186/s13059-021-02558-x ◽

2021 ◽

Vol 22 (1) ◽

Author(s):

Michael M. Khayat ◽

Sayed Mohammad Ebrahim Sahraeian ◽

Samantha Zarate ◽

Andrew Carroll ◽

Huixiao Hong ◽

...

Keyword(s):

Next Generation Sequencing ◽

False Negative ◽

False Negative Rate ◽

Next Generation Sequencing Data ◽

Chinese Family ◽

Next Generation ◽

Sequencing Data ◽

Structural Variations ◽

The Impact ◽

Generation Sequencing

Abstract Background Genomic structural variations (SV) are important determinants of genotypic and phenotypic changes in many organisms. However, the detection of SV from next-generation sequencing data remains challenging. Results In this study, DNA from a Chinese family quartet is sequenced at three different sequencing centers in triplicate. A total of 288 derivative data sets are generated utilizing different analysis pipelines and compared to identify sources of analytical variability. Mapping methods provide the major contribution to variability, followed by sequencing centers and replicates. Interestingly, SV supported by only one center or replicate often represent true positives with 47.02% and 45.44% overlapping the long-read SV call set, respectively. This is consistent with an overall higher false negative rate for SV calling in centers and replicates compared to mappers (15.72%). Finally, we observe that the SV calling variability also persists in a genotyping approach, indicating the impact of the underlying sequencing and preparation approaches. Conclusions This study provides the first detailed insights into the sources of variability in SV identification from next-generation sequencing and highlights remaining challenges in SV calling for large cohorts. We further give recommendations on how to reduce SV calling variability and the choice of alignment methodology.

Download Full-text

Diagnostic Accuracy and Clinical Impact of Sentinel Lymph Node Sampling in Endometrial Cancer at High Risk of Recurrence: A Meta-Analysis

Journal of Clinical Medicine ◽

10.3390/jcm9123874 ◽

2020 ◽

Vol 9 (12) ◽

pp. 3874

Author(s):

Lise Lecointre ◽

Massimo Lodi ◽

Émilie Faller ◽

Thomas Boisramé ◽

Vincent Agnus ◽

...

Keyword(s):

Clinical Trials ◽

Endometrial Cancer ◽

Lymph Node ◽

High Risk ◽

Sentinel Lymph Node ◽

Meta Analysis ◽

False Negative ◽

False Negative Rate ◽

Number Of Patients ◽

The Impact

Purpose. To assess the value of sentinel lymph node (SLN) sampling in high risk endometrial cancer according to the ESMO-ESGO-ESTRO classification. Methods. We performed a comprehensive search on PubMed for clinical trials evaluating SLN sampling in patients with high risk endometrial cancer: stage I endometrioid, grade 3, with at least 50% myometrial invasion, regardless of lymphovascular space invasion status; or stage II; or node-negative stage III endometrioid, no residual disease; or non-endometrioid (serous or clear cell or undifferentiated carcinoma, or carcinosarcoma). All patients underwent SLN sampling followed by pelvic with or without para-aortic lymphadenectomy. Results. We included 17 original studies concerning 1322 women. Mean detection rates were 89% for unilateral and 68% for bilateral. Pooled sensitivity was 88.5% (95%CI: 81.2–93.2%), negative predictive value was 96.0% (95%CI: 93.1–97.7%), and false negative rate was 11.5% (95%CI: 6.8; 18.8%). We noted heterogeneity in SLN techniques between studies, concerning the tracer and its detection, the injection site, the number of injections, and the surgical approach. Finally, we found a correlation between the number of patients included and the SLN sampling performances. Discussion. This meta-analysis estimated the SLN sampling performances in high risk endometrial cancer patients. Data from the literature show the feasibility, the safety, the limits, and the impact on surgical de-escalation of this technique. In conclusion, our study supports the hypothesis that SLN sampling could be a valuable technique to diagnose lymph node involvement for patients with high risk endometrial cancer in replacement of conventional lymphadenectomy. Consequently, randomized clinical trials are necessary to confirm this hypothesis.

Download Full-text

Effect of Sample Aliquot Size on the Limit of Detection and Reproducibility of Clinical Assays

Clinical Chemistry ◽

10.1373/clinchem.2007.089854 ◽

2007 ◽

Vol 53 (11) ◽

pp. 1962-1965 ◽

Cited By ~ 6

Author(s):

Guorong Chen ◽

Lori Kobayashi ◽

Irina Nazarenko

Keyword(s):

Limit Of Detection ◽

Probability Model ◽

False Negative ◽

Success Probability ◽

False Negative Rate ◽

Sample Volume ◽

Nucleic Acid Amplification ◽

Clinical Samples ◽

Minimum Number ◽

The Impact

Abstract Background: Nucleic acid amplification technologies significantly improved the limit of detection (LOD) for diagnostic assays. The ability of these assays to amplify fewer than 10 target copies of DNA or RNA imposes new requirements on the preparation of clinical samples. We report a statistical method to determine how large of an aliquot is necessary to reproducibly provide a detectable number of cells. Methods: We determined the success probability (p) based on aliquot size and sample volume. The binomial distribution, based on p and the concentration of cells in sample, was used to calculate the probability of getting no target objects in an aliquot and to determine the minimum number of objects per aliquot necessary to generate a reproducible clinical assay. Results: The described method was applied to find a minimum aliquot volume required for a set LOD, false-negative rate (FNR), and %CV. For example, to keep FNR <0.01% for 0.5%, 1% and 2% aliquots (minimum 2000, 1000, and 500 cells per sample) are required. Comparison between experimental and predicted FNR demonstrated good correlation for the small volume aliquots and/or low concentration of target. When 4 μL of 200 copies/mL of plasmid is amplified, predicted and experimental FNRs are 47.2% and 44.9%. Conclusion: This probability model is a useful tool to predict the impact of aliquot volume on the LOD and reproducibility of clinical assays. Even for samples for which pathogens are homogeneously distributed, it is theoretically impossible to collect a single pathogen consistently if the concentration of pathogen is below a certain limit.

Download Full-text

Accuracy of Healthcare Professionals’ Nasopharyngeal Swab Technique in SARS-CoV-2 Specimen Collection

10.1101/2020.10.19.20213140 ◽

2020 ◽

Author(s):

Robbie S. R. Woods ◽

Kellie Nwaokorie ◽

Jana Crowley ◽

Michael Walsh ◽

Eoghan de Barra ◽

...

Keyword(s):

Healthcare Professionals ◽

False Negative ◽

Three Dimensional ◽

False Negative Rate ◽

Nasopharyngeal Swab ◽

Three Dimensional Model ◽

Tertiary Referral Centre ◽

Specimen Collection ◽

A Prospective Study ◽

Novel Design

AbstractBackgroundThe COVID-19 pandemic has caused huge pressure on healthcare systems worldwide. Public health measures to control the virus are reliant on testing, including appropriate collection of specimens for analysis.MethodsA prospective study of nasopharyngeal swab technique by staff in an academic tertiary referral centre was carried out. Nasopharyngeal swab technique was evaluated by a novel design of a navigated swab on a three-dimensional model head.ResultsSwab technique of 228 participants was assessed. Technique was poor, with a success rate of nasopharyngeal swabbing at 38.6%. Angle and length of insertion were significantly different between those with successful and unsuccessful technique. Doctors were significantly more accurate than nurses and non-healthcare professionals (p<0.01).ConclusionInaccurate specimen collection from poor swab technique could contribute to a false negative rate of testing for SARS-CoV-2. Specific training in nasopharyngeal anatomy and swab technique may improve the accuracy of nasopharyngeal swabbing.

Download Full-text

The Effect of a Diverse Dataset for Transfer Learning in Thermal Person Detection

Sensors ◽

10.3390/s20071982 ◽

2020 ◽

Vol 20 (7) ◽

pp. 1982 ◽

Cited By ~ 1

Author(s):

Noor Ul Huda ◽

Bolette D. Hansen ◽

Rikke Gade ◽

Thomas B. Moeslund

Keyword(s):

Transfer Learning ◽

Privacy Preservation ◽

False Negative ◽

False Negative Rate ◽

Model Adaptation ◽

True Negative ◽

True Negative Rate ◽

Negative Rate ◽

Thermal Cameras ◽

The Impact

Thermal cameras are popular in detection for their precision in surveillance in the dark and for privacy preservation. In the era of data driven problem solving approaches, manually finding and annotating a large amount of data is inefficient in terms of cost and effort. With the introduction of transfer learning, rather than having large datasets, a dataset covering all characteristics and aspects of the target place is more important. In this work, we studied a large thermal dataset recorded for 20 weeks and identified nine phenomena in it. Moreover, we investigated the impact of each phenomenon for model adaptation in transfer learning. Each phenomenon was investigated separately and in combination. the performance was analyzed by computing the F1 score, precision, recall, true negative rate, and false negative rate. Furthermore, to underline our investigation, the trained model with our dataset was further tested on publicly available datasets, and encouraging results were obtained. Finally, our dataset was also made publicly available.

Download Full-text

Transfer Detection of YOLO to Focus CNN’s Attention on Nude Regions for Adult Content Detection

Symmetry ◽

10.3390/sym13010026 ◽

2020 ◽

Vol 13 (1) ◽

pp. 26

Author(s):

Nouar AlDahoul ◽

Hezerul Abdul Karim ◽

Mohd Haris Lye Abdullah ◽

Mohammad Faizal Ahmad Fauzi ◽

Abdulaziz Saleh Ba Wazir ◽

...

Keyword(s):

Recent Literature ◽

False Negative ◽

False Negative Rate ◽

Regions Of Interest ◽

Support Vector ◽

Small Scale ◽

High False Negative Rate ◽

Diverse Backgrounds ◽

Ablation Study ◽

The Impact

Video pornography and nudity detection aim to detect and classify people in videos into nude or normal for censorship purposes. Recent literature has demonstrated pornography detection utilising the convolutional neural network (CNN) to extract features directly from the whole frames and support vector machine (SVM) to classify the extracted features into two categories. However, existing methods were not able to detect the small-scale content of pornography and nudity in frames with diverse backgrounds. This limitation has led to a high false-negative rate (FNR) and misclassification of nude frames as normal ones. In order to address this matter, this paper explores the limitation of the existing convolutional-only approaches focusing the visual attention of CNN on the expected nude regions inside the frames to reduce the FNR. The You Only Look Once (YOLO) object detector was transferred to the pornography and nudity detection application to detect persons as regions of interest (ROIs), which were applied to CNN and SVM for nude/normal classification. Several experiments were conducted to compare the performance of various CNNs and classifiers using our proposed dataset. It was found that ResNet101 with random forest outperformed other models concerning the F1-score of 90.03% and accuracy of 87.75%. Furthermore, an ablation study was performed to demonstrate the impact of adding the YOLO before the CNN. YOLO–CNN was shown to outperform CNN-only in terms of accuracy, which was increased from 85.5% to 89.5%. Additionally, a new benchmark dataset with challenging content, including various human sizes and backgrounds, was proposed.

Download Full-text

A Novel Compartmental Model for Analysis and Projection of COVID-19 Dynamics in Bangladesh

Advances in Research ◽

10.9734/air/2020/v21i930228 ◽

2020 ◽

pp. 14-28

Author(s):

Md. Mijanur Rahman ◽

Md. Sadekur Rahman Rani

Keyword(s):

Compartmental Model ◽

False Negative ◽

False Negative Rate ◽

Model Parameters ◽

Contact Rate ◽

Rate Analysis ◽

Negative Rate ◽

Exposed Population ◽

Effective Contact ◽

The Impact

A novel compartmental model is proposed to project the COVID-19 dynamics in Bangladesh. The exposed population is divided into two classes: tested and not tested. Model parameters are estimated by fitting the output with empirical COVID-19 data of Bangladesh from 7 April 2020 to 15 June 2020. It is found that even if 90% of exposed individuals are tested, number of unidentified cases (recovered or dead) is 3 to 4 times than that of identified cases. As of 15 June 2020, Bangladesh is using the Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) based test to detect the novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). The impact of false negative rate of this test on unidentified infection is analyzed. It is found that the year-end total recoveries (deaths) surges 700 (800) times if the false negative rate is doubled. Periodic lockdown and relaxation intervals are incorporated by defining the effective contact rate (β) as a periodic function of time. Impact of lockdown is perspicuous from the periodic fluctuation of the basic reproduction number ( ). It is observed that a 90-day-lockdwon reduces the final outcome by 3% while a 30-day-lockdwon increases it by 2%. On other hand, casualties are 10 to 100 times worse in case of no lockdown even with less than half effective contact rate. Analysis of strictness of isolation reveals that a 12.5% increase in the strictness coefficient reduces the exposed population 2.5 times whereas a 37.5% decrease in it intensifies the outcome nearly 9 times. Projections up to 6 April 2021 suggests that the epidemic will reach its peak in Bangladesh in August 2020.

Download Full-text