Predicting Kidney Graft Survival Using Machine Learning Methods: Prediction Model Development and Feature Significance Analysis Study.

BACKGROUND: Kidney transplantation is the optimal treatment for patients with end-stage renal disease. Short- and long-term kidney graft survival is influenced by a number of donor and recipient factors. Predicting the success of kidney transplantation is important for optimizing kidney allocation. OBJECTIVE: The aim of this study was to predict the risk of kidney graft failure across three temporal cohorts (within 1 year, within 5 years, and after 5 years following a transplant) based on donor and recipient characteristics. We analyzed a large data set comprising over 50,000 kidney transplants covering an approximate 20-year period. METHODS: We applied machine learning-based classification algorithms to develop prediction models for the risk of graft failure for three different temporal cohorts. Deep learning-based autoencoders were applied for data dimensionality reduction, which improved the prediction performance. The influence of features on graft survival for each cohort was studied by investigating a new nonoverlapping patient stratification approach. RESULTS: Our models predicted graft survival with area under the curve scores of 82% within 1 year, 69% within 5 years, and 81% within 17 years. The feature importance analysis elucidated the varying influence of clinical features on graft survival across the three different temporal cohorts. CONCLUSIONS: In this study, we applied machine learning to develop risk prediction models for graft failure that demonstrated a high level of prediction performance. Acknowledging that these models performed better than those reported in the literature for existing risk prediction tools, future studies will focus on how best to incorporate these prediction models into clinical care algorithms to optimize the long-term health of kidney recipients.

Polyclonal antibodies selectively inhibit tumor growth and invasion and synergize with immune checkpoint inhibitors.

Heterologous polyclonal antibodies (pAb) were shown to possess oncolytic properties a century ago with reported clinical responses. More recent preclinical models confirmed pAb efficacy, though their ability to tackle complex target antigens reduces susceptibility to tumor escape. Owing to the recent availability of glyco-humanized pAb (GH-pAb) with acceptable clinical toxicology profile, we revisited use of pAb in oncology and highlighted their therapeutic potential against multiple cancer types. Murine antitumor pAb were generated after repeated immunization of rabbits with murine tumor cell lines from hepatocarcinoma, melanoma, and colorectal cancers. Antitumor pAb recognized and showed cytotoxicity against their targets without cross-reactivity with healthy tissues. In vivo, pAb are effective alone; moreover, these pAb synergize with immune checkpoint inhibitors like anti-PD-L1 in several cancer models. They elicited an antitumor host immune response and prevented metastases. The anticancer activity of pAb was also confirmed in xenografted NMRI nude mice using GH-pAb produced by repeated immunization of pigs with human tumor cell lines. In conclusion, the availability of bioengineered GH-pAb allows for revisiting of passive immunotherapy with oncolytic pAb to fight against solid tumor and cancer metastasis.

Gait biomechanics phenotypes among total knee arthroplasty candidates by machine learning cluster analysis.

Knee osteoarthritis patient phenotyping is relevant to developing targeted treatments and assessing the treatment efficacy of total knee arthroplasty (TKA). This study aimed to identify clusters among TKA candidates based on demographic and knee mechanic features during gait, and compare gait changes between clusters postoperatively. TKA patients underwent 3D gait analysis 1-week pre (n = 134) and 1-year post-TKA (n = 105). Principal component analysis was applied to frontal and sagittal knee angle and moment waveforms, extracting major patterns of variability. Age, sex, body mass index, gait speed, and frontal and sagittal pre-TKA angle and moment PC scores previously identified as relevant to TKA outcomes were standardized (mean = 0, SD = 1, [134 × 15]). Multidimensional scaling and machine learning-based hierarchical clustering were applied. Final clusters were validated by examining intercluster differences pre-TKA and gait feature changes (PostPCscore - PrePCscore ) by k-way Χ2 and ANOVA tests. Four TKA candidate phenotypes yielded optimum clustering metrics, interpreted as higher and lower functioning clusters that were predominantly male and female. Higher functioning clusters pre-TKA (clusters 1 and 4) had more dynamic sagittal flexion moment (p < 0.001) and frontal plane adduction moment (p < 0.001) loading/un-loading patterns during stance. Post-TKA, higher functioning clusters demonstrated less knee mechanic improvements during gait (flexion angle p < 0.001; flexion moment p < 0.001). TKA candidates can be characterized by four clusters, predominately separated by sex and knee joint biomechanics. Post-TKA knee kinematics and kinetics improvements were cluster-specific; lower functioning clusters experienced more improvement. Cluster-based patient profiling may aid in triaging and developing OA management and surgical strategies meeting group-level function needs.

Effectiveness and Safety of Tofacitinib in Canadian Patients With Rheumatoid Arthritis: Primary Results From a Prospective Observational Study.

OBJECTIVE: The Canadian Tofacitinib for Rheumatoid Arthritis Observational (CANTORAL) is the first Canadian prospective, observational study assessing tofacitinib. The objective was to assess effectiveness and safety for moderate to severe rheumatoid arthritis (RA). Coprimary and secondary outcomes are reported from an interim analysis. METHODS: Patients initiating tofacitinib from October 2017 to July 2020 were enrolled from 45 Canadian sites. Coprimary outcomes (month 6) included the Clinical Disease Activity Index (CDAI)-defined low disease activity (LDA) and remission. Secondary outcomes (to month 18) included the CDAI and the 4-variable Disease Activity Score in 28 joints (DAS28) using the erythrocyte sedimentation rate (ESR)/C-reactive protein (CRP) level to define LDA and remission; the proportions of patients achieving mild pain (visual analog scale <20 mm), and moderate (≥30%) and substantial (≥50%) pain improvements; and the proportions of patients achieving a Health Assessment Questionnaire disability index (HAQ DI) score greater or equal to normative values (≤0.25) and a HAQ DI score greater or equal to minimum clinically important difference (MCID) (≥0.22). Safety was assessed to month 36. RESULTS: Of 504 patients initiating tofacitinib, 44.4% received concomitant methotrexate. At month 6, 52.9% and 15.4% of patients were in CDAI-defined LDA and remission, respectively; a similar proportion of patients achieved outcomes by month 3 (first post-baseline assessment). By month 3, 27.2% and 41.7% of patients, respectively, were in DAS28-ESR-defined LDA and DAS28-CRP <3.2; 14.7% and 25.8% achieved DAS28-ESR remission and DAS28-CRP <2.6. By month 3, mild pain and moderate and substantial pain improvements occurred in 29.6%, 55.6%, and 42.9% of patients, respectively; 19.9% and 53.7% of patients achieved a HAQ DI score greater than or equal to normative values and a HAQ DI score greater than or equal to MCID, respectively. Outcomes were generally maintained to month 18. Incidence rates (events per 100 patient-years) for treatment-emergent adverse events (AEs), serious AEs, and discontinuations due to AEs were 126.8, 11.9, and 14.5, respectively, and AEs of special interest were infrequent. CONCLUSION: Tofacitinib was associated with early and sustained improvement in RA signs and symptoms in real-world patients. Effectiveness and safety were consistent with the established tofacitinib clinical profile.

Indoor location identification of patients for directing virtual care: An AI approach using machine learning and knowledge-based methods.

In a digitally enabled healthcare setting, we posit that an individual's current location is pivotal for supporting many virtual care services-such as tailoring educational content towards an individual's current location, and, hence, current stage in an acute care process; improving activity recognition for supporting self-management in a home-based setting; and guiding individuals with cognitive decline through daily activities in their home. However, unobtrusively estimating an individual's indoor location in real-world care settings is still a challenging problem. Moreover, the needs of location-specific care interventions go beyond absolute coordinates and require the individual's discrete semantic location; i.e., it is the concrete type of an individual's location (e.g., exam vs. waiting room; bathroom vs. kitchen) that will drive the tailoring of educational content or recognition of activities. We utilized Machine Learning methods to accurately identify an individual's discrete location, together with knowledge-based models and tools to supply the associated semantics of identified locations. We considered clustering solutions to improve localization accuracy at the expense of granularity; and investigate sensor fusion-based heuristics to rule out false location estimates. We present an AI-driven indoor localization approach that integrates both data-driven and knowledge-based processes and artifacts. We illustrate the application of our approach in two compelling healthcare use cases, and empirically validated our localization approach at the emergency unit of a large Canadian pediatric hospital.

A Self-Setting Hydrogel of Silylated Chitosan and Cellulose for the Repair of Osteochondral Defects: From in vitro Characterization to Preclinical Evaluation in Dogs.

Articular cartilage (AC) may be affected by many injuries including traumatic lesions that predispose to osteoarthritis. Currently there is no efficient cure for cartilage lesions. In that respect, new strategies for regenerating AC are contemplated with interest. In this context, we aim to develop and characterize an injectable, self-hardening, mechanically reinforced hydrogel (Si-HPCH) composed of silanised hydroxypropymethyl cellulose (Si-HPMC) mixed with silanised chitosan. The in vitro cytocompatibility of Si-HPCH was tested using human adipose stromal cells (hASC). In vivo, we first mixed Si-HPCH with hASC to observe cell viability after implantation in nude mice subcutis. Si-HPCH associated or not with canine ASC (cASC), was then tested for the repair of osteochondral defects in canine femoral condyles. Our data demonstrated that Si-HPCH supports hASC viability in culture. Moreover, Si-HPCH allows the transplantation of hASC in the subcutis of nude mice while maintaining their viability and secretory activity. In the canine osteochondral defect model, while the empty defects were only partially filled with a fibrous tissue, defects filled with Si-HPCH with or without cASC, revealed a significant osteochondral regeneration. To conclude, Si-HPCH is an injectable, self-setting and cytocompatible hydrogel able to support the in vitro and in vivo viability and activity of hASC as well as the regeneration of osteochondral defects in dogs when implanted alone or with ASC.

A Data Mining Framework for Glaucoma Decision Support Based on Optic Nerve Image Analysis Using Machine Learning Methods.

Ocular imaging instruments, such as Confocal Scanning Laser Ophthalmoscopy (CSLO), captures high-quality images of the optic disc (also known as optic nerve head) that help clinicians to diagnose glaucoma. We present an integrated data analytics framework to aid clinicians in interpreting CSLO optic nerve images to diagnose and monitor the progression of glaucoma. To distinguish between healthy and glaucomatous optic discs, our framework derives shape information from CSLO images using image processing (Zernike moment method), selects salient features (hybrid feature selection), and then trains image classifiers (Multilayer Perceptron, Support Vector Machine, Bayesian Network). To monitor glaucoma progression over time, our framework uses a mathematical model of the optic disc to extract morphological features from CSLO images and applies clustering (Self-Organizing Maps) to visualize subtypes of glaucomatous optic disc damage. We contend that our data analytics framework offers an automated and objective analysis of optic nerve images that can potentially support both diagnosis and monitoring of glaucoma. We validated our framework with CSLO optic nerve images and our data analytics approach detected glaucomatous optic discs with a sensitivity of 0.86, a specificity of 0.80, an accuracy of 0.838, and an AUROC of 0.913 with a Bayesian network classifier using the optimal subset of Zernike features (six moments). Furthermore, our framework identified, using morphological features, five clusters of CSLO images, where each cluster stands for a subtype of optic nerve damage (two healthy subtypes and three glaucoma subtypes). The characteristics of each cluster-the subtype of the image-were determined by experts who examined the morphology of the images within each cluster and provided subtype characteristics to each cluster.

Monitoring Activities Related to Medication Adherence in Ambient Assisted Living Environments.

A recent trend in healthcare is to motivate patients to self-manage their health conditions in home-based settings. Medication adherence is an important aspect in disease self-management since sub-optimal medication adherence by the patient can lead to serious healthcare costs and discomfort for the patient. In order to alleviate the limitations of self-reported medication adherence, we can use ambient assistive living (AAL) technologies in smart environments. Activity recognition services allow to retrieve self-management information related to medication adherence in a less intrusive way. By remotely monitor compliance with medication adherence, self-management program's interventions can be tailored and adapted based on the observed patient's behaviour. To address this challenge, we present an AAL framework that monitor activities related to medication adherence.

A Mobile and Intelligent Patient Diary for Chronic Disease Self-Management.

By involving patients in their own long-term care, patient self-management approaches aim to increase self-sufficiency and reduce healthcare costs. For example, electronic patient diaries enable patients to collect health data autonomously, increasing self-reliance and reducing strain on health professionals. By deploying patient diaries on mobile platforms, health data collection can occur at any time and place, increasing the mobility of chronic patients who typically need to enter health data frequently. Importantly, an opportunity also arises for mobile clinical decision support, where health feedback is directly issued to patients without relying on connectivity or remote servers. Regardless of the specific self-management strategy, patient and healthcare provider adoption are crucial. Tailoring the system towards the particular patient and toward institution-specific clinical pathways is essential to increasing acceptance. In this paper we discuss a mobile patient diary realizing both the opportunities and challenges of mobile deployment.

Metatarsal bars more effective than metatarsal pads in reducing impulse on the second metatarsal head.

BACKGROUND: The second metatarsal head is commonly involved in cases of metatarsalgia. As part of the conservative treatment, metatarsal bars and metatarsal pads are often prescribed. OBJECTIVE: To compare the effectiveness of metatarsal bars and metatarsal pads in reducing impulse on the second metatarsal head. METHOD: Thirty-five healthy subjects were monitored with an insole scanning system during walking in four different conditions: (a) wearing shoes only, (b) shoes plus metatarsal pads and shoes plus metatarsal bars, placed either (c) perpendicular to the foot axis or (d) oblique to the foot axis. The impulse under the second metatarsal head was measured using the first condition as a control. Both feet were examined in each subject resulting in a total of 840 measurements. RESULTS: Both metatarsal bars and metatarsal pads were effective in reducing impulse when compared with the control (P<0.01). Metatarsal bars were found to be more effective in reducing impulse as compared to the metatarsal pads (P<0.01), and the oblique position of the bars was more effective than the perpendicular one (P<0.01). CONCLUSIONS: The greatest reduction of impulse on the second metatarsal head in healthy subjects is achieved with the use of metatarsal bars in an oblique position.