A Survey on Concept Drift in Process Mining

Concept drift in process mining (PM) is a challenge as classical methods assume processes are in a steady-state, i.e., events share the same process version. We conducted a systematic literature review on the intersection of these areas, and thus, we review concept drift in PM and bring forward a taxonomy of existing techniques for drift detection and online PM for evolving environments. Existing works depict that (i) PM still primarily focuses on offline analysis, and (ii) the assessment of concept drift techniques in processes is cumbersome due to the lack of common evaluation protocol, datasets, and metrics.

Malignancy prediction of cutaneous and subcutaneous neoplasms in canines using B-mode ultrasonography, Doppler, and ARFI elastography

Background Cutaneous and subcutaneous neoplasms are highly prevalent in dogs, ranging from benign to highly aggressive and metastatic lesions. The diagnosis is obtained through histopathology, however it is an invasive technique that may take a long time to obtain the result, delaying the beginning of the adequate treatment. Thus, there is a need for non-invasive tests that can help in the early diagnosis of this type of cancer. The aim of this study was to verify the accuracy of B-mode ultrasonography, Doppler, and ARFI elastography to predict malignancy in cutaneous and subcutaneous canine neoplasms. In addition, we aim to propose an ultrasonography evaluation protocol and perform the neoplasms characterization using these three proposed techniques. Results Twenty-one types of specific neoplasm were diagnosed, and using B-mode, we verified the association between heterogeneous echotexture, invasiveness, presence of hyperechoic spots, and cavity areas with malignancy. An increased pulsatility was verified in malignant neoplasms using Doppler (cut-off value > 0.93). When using the elastography, malignancy was associated with non-deformable tissues and shear wave velocity > 3.52 m/s. Evaluation protocols were proposed associating 4, 5, 6, or 7 malignancy predictive characteristics, and characterization was done for all tumors with at least two cases. Conclusions We concluded that ultrasonography methods are promising and effective in predicting malignancy in these types of tumors, and the association of methods can increase the specificity of the results.

Setting Up the Stillbirth Registration System and Investigating the Related Causes in Iran in Selected Hospitals: Observational Cohort Study

Background One of the major causes of perinatal mortality is stillbirth. In many cases, the cause of stillbirth is difficult to identify, and the cause of many cases remain unexplained. Because of the lack of registration stillbirth system in our country we developed protocol and instructions for stillbirth and setting up a stillbirth registration system in selected hospitals around country. Methods Iranian Maternal and Neonatal Network (IMaN) registers information about almost all births (live & dead) around the country, but this network does not collect data about stillbirth causes. In this study, we developed the stillbirth evaluation protocol with experts' cooperation, and we designed forms for the stillbirth registration system electronically. Then we trained related individuals in 14 selected hospital from 12 provinces (14 cities) of Iran. After a year, we extracted, analyzed, and, based on the Relevant Condition of Death Classification (ReCoDe), interpreted the collected data. Results A total of 105,562 births and 762 stillbirths registered. In 742 registered stillbirth cases in 14 selected hospitals, the relevant causes were identified in 65.4% of cases, while 34.6% of cases remained unclassified. The most frequent relevant conditions were fetal (33.2%), maternal (9.1%), amniotic fluid (8.8%), placenta (7.7%), and umbilical cord (6.2%). Conclusions Our registration decreased the percentage of stillbirth with an unexplained cause from about 70–34.6%.

TractoInferno: A large-scale, open-source, multi-site database for machine learning dMRI tractography

TractoInferno is the world's largest open-source multi-site tractography database, including both research- and clinical-like human acquisitions, aimed specifically at machine learning tractography approaches and related ML algorithms. It provides 284 datasets acquired from 3T scanners across 6 different sites. Available data includes T1-weighted images, single-shell diffusion MRI (dMRI) acquisitions, spherical harmonics fitted to the dMRI signal, fiber ODFs, and reference streamlines for 30 delineated bundles generated using 4 tractography algorithms, as well as masks needed to run tractography algorithms. Manual quality control was additionally performed at multiple steps of the pipeline. We showcase TractoInferno by benchmarking the learn2track algorithm and 5 variations of the same recurrent neural network architecture. Creating the TractoInferno database required approximately 20,000 CPU-hours of processing power, 200 man-hours of manual QC, 3,000 GPU-hours of training baseline models, and 4 Tb of storage, to produce a final database of 350 Gb. By providing a standardized training dataset and evaluation protocol, TractoInferno is an excellent tool to address common issues in machine learning tractography.

Efficient protocols for oblivious linear function evaluation from ring-LWE1

An oblivious linear function evaluation protocol, or OLE, is a two-party protocol for the function f ( x ) = a x + b, where a sender inputs the field elements a, b, and a receiver inputs x and learns f ( x ). OLE can be used to build secret-shared multiplication, and is an essential component of many secure computation applications including general-purpose multi-party computation, private set intersection and more. In this work, we present several efficient OLE protocols from the ring learning with errors (RLWE) assumption. Technically, we build two new passively secure protocols, which build upon recent advances in homomorphic secret sharing from (R)LWE (Boyle et al. in: EUROCRYPT 2019, Part II (2019) 3–33 Springer), with optimizations tailored to the setting of OLE. We upgrade these to active security using efficient amortized zero-knowledge techniques for lattice relations (Baum et al. in: CRYPTO 2018, Part II (2018) 669–699 Springer), and design new variants of zero-knowledge arguments that are necessary for some of our constructions. Our protocols offer several advantages over existing constructions. Firstly, they have the lowest communication complexity amongst previous, practical protocols from RLWE and other assumptions; secondly, they are conceptually very simple, and have just one round of interaction for the case of OLE where b is randomly chosen. We demonstrate this with an implementation of one of our passively secure protocols, which can perform more than 1 million OLEs per second over the ring Z m , for a 120-bit modulus m, on standard hardware.

Characterization of Cutaneous and Subcutaneous Neoplasms in Canines and Malignancy Prediction Using B-Mode Ultrasonography, Doppler, and ARFI Elastography

Background Cutaneous and subcutaneous neoplasms are highly prevalent in dogs, ranging from benign to highly aggressive and metastatic lesions. The diagnosis is obtained through histopathology, however it is an invasive technique that may take a long time to obtain the result, delaying the beginning of the adequate treatment. Thus, there is a need for non-invasive tests that can help in the early diagnosis of this type of cancer. The aim of this study was to verify the accuracy of ultrasonography methods to predict malignancy in cutaneous and subcutaneous canine neoplasms. In addition, we aim to propose an ultrasonography evaluation protocol and perform the neoplasms characterization using these three proposed techniques. Results Twenty-one types of specific neoplasm were diagnosed, and using B-mode, we verified the association between heterogeneous echotexture, invasiveness, presence of hyperechoic spots, and cavity areas with malignancy. An increased pulsatility was verified in malignant neoplasms using Doppler (cut-off value >0.93). When using the elastography, malignancy was associated with non-deformable tissues and shear wave velocity >3.52 m/s. Evaluation protocols were proposed associating 4, 5, 6, or 7 malignancy predictive characteristics, and characterization was done for all tumors with at least two cases. Conclusions We concluded that ultrasonography methods are promising and effective in predicting malignancy in these types of tumors, and the association of methods can increase the specificity of the results.

A new digital evaluation protocol applied in a retrospective analysis of periodontal plastic surgery of gingival recessions

This research aimed to develop a new digital evaluation protocol to objectively quantify the volumetric changes of root coverage periodontal plastic surgery when combined with connective tissue graft. Consecutive patients with Cairo recession type 1 (RT1) or Cairo recession type 2 (RT2) were treated. Accurate study models obtained at baseline and follow-ups were optically scanned. Healing dynamics were measured by calculating volume differences between time points. Nineteen patients were treated between December 2014 and January 2019. At 3-month follow-up, root coverage was 95.6% (± 14.5%) with tunnel and connective tissue graft (TUN + CTG) technique, and 88.9% (± 20.5%) with the vestibular incision subperiosteal tunnel access and connective tissue graft (VISTA + CTG) technique. Recession decreased 1.33 (± 0.86) mm and 1.42 (± 0.92) mm, respectively (p = 0.337). At 6-month follow-up, root coverage was 96.5% (± 10.4%) with the TUN + CTG and 93.9% (± 10.3%) with the VISTA + CTG. Recession decreased 1.35 (± 0.85) mm and 1.45 (± 0.82) mm, respectively (p = 0.455). Complete root coverage was achieved in 86.7% (± 0.4%) with TUN + CTG and 70.6% (± 0.5%) with VISTA + CTG. No statistically significant differences were found between techniques. The digital protocol presented proved to be a non-invasive technique for accurate measurements of clinical outcomes. Both techniques reduce gingival recessions, with no statistically significant differences.