WKWilliam Klement

William Klement

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


William Klement has a total of 46 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Chest Tube Learning Synthesis and Evaluation Assistant (CheLSEA): A Prospective Observational Trial of an Intelligent Decision Support System

    by Nikita Arora, William Klement, Nathalie Japkowicz, Daniel G. Jones, Donna E. Maziak, Andrew J.E. Seely, Sudhir R. Sundaresan, Patrick J. Villeneuve, Sebastien Gilbert

    Objective: This study evaluates the performance of an artificial intelligence predictive clinical decision support system (CheLSEA) in generating chest tube management recommendations. Methods: From October 2020 to May 2021, 50 adult elective pulmonary resection patients with at least 24 h of chest tube drainage were enrolled in a single-arm, double-anonymized, observational study to evaluate CheLSEA’s performance compared with standard chest tube care. Clinical status, digital pleural drainage data, and chest X-ray data were collected prospectively. For each query, CheLSEA generated a recommendation for chest tube removal or maintenance. If maintenance was recommended, CheLSEA generated a removal time prediction. Results: Most patients were female (29 of 47, 62%), smokers (39 of 47, 83%), with a median age of 73 (interquartile range [IQR]: 66 to 77) years, who underwent minimally invasive (44 of 47, 94%) lobectomy (41 of 47, 87%) for primary non-small cell lung cancer (35 of 47, 75%). CheLSEA was queried 174 times, 21% (36 of 174) of which triggered the CheLSEA safeguard system, mostly due to grade 3 or increasing subcutaneous emphysema (20 of 36, 56%). CheLSEA recommended chest tube removal in 9% of remaining requests (12 of 138), 83% of which were safe (10 of 12) and 17% of which were premature by ≤6 h (2 of 12). The remaining 126 queries were answered with chest tube maintenance recommendations up to the optimal removal time (97 of 126, 77%) or shortly thereafter (29 of 126, 23%; median = 17 h, IQR: 17 to 22). When predicting chest tube removal time, 93% of responses (82 of 88) were accurate. Conclusions: CheLSEA provides safe chest tube management recommendations and can potentially enhance care by reliably emulating expert-level clinical guidance.

  • Hospital-Acquired and Ventilator-Associated Pneumonia Early After Lung Transplantation: A Prospective Study on Incidence, Pathogen Origin, and Outcome

    by Laura N Walti, Chun Fai Ng, Qasim Mohiuddin, Roni Bitterman, Mohammed Alsaeed, William Klement, Tereza Martinu, Aman Sidhu, Tony Mazzulli, Laura Donahoe, et al.

    Background: Hospital-acquired (HAP) and ventilator-associated pneumonia (VAP) are important complications early (<30 days) after lung transplantation (LT). However, current incidence, associated factors, and outcomes are not well reported. Methods: LT recipients transplanted at our institution (July 2019–January 2020 and October 2021–November 2022) were prospectively included. We assessed incidence and presentation of pneumonia and evaluated the impact of associated factors using regression models. We also evaluated molecular relatedness of respiratory pathogens collected peri-transplant and at pneumonia occurrence using pulsed-field gel electrophoresis (PFGE). Results: In the first 30 days post-LT, 25/270 (9.3%) recipients were diagnosed with pneumonia (68% [17/25] VAP; 32% [8/25] HAP). Median time to pneumonia was 11 days (IQR, 7–13); 49% (132/270) of donor and 16% (44/270) of recipient respiratory peri-transplant cultures were positive. However, pathogens associated with pneumonia were not genetically related to either donor or recipient cultures at transplant, as determined by PFGE. Diagnosed pulmonary hypertension (HR, 4.42; 95% CI, 1.62–12.08) and immunosuppression use (HR, 2.87; 95% CI, 1.30–6.56) were pre-transplant factors associated with pneumonia. Pneumonia occurrence was associated with longer hospital stay (HR, 5.44; 95% CI, 2.22–13.37) and VAP with longer ICU stay (HR, 4.31; 95% CI, 1.73–10.75) within the first 30 days post-transplantation; 30- and 90-day mortality were similar. Conclusions: Prospectively assessed early pneumonia incidence occurred in ∼10% of LT. Populations at increased risk for pneumonia occurrence include LT with pre-transplant pulmonary hypertension and pre-transplant immunosuppression. Pneumonia was associated with increased healthcare use, highlighting the need for further improvements by preferentially targeting higher-risk patients.


  • Open Access

    Consolidated Reporting Guidelines for Prognostic and Diagnostic Machine Learning Models (CREMLS)

    by Khaled El Emam, Tiffany I Leung, Bradley Malin, William Klement, Gunther Eysenbach

    The number of papers presenting machine learning (ML) models that are being submitted to and published in the Journal of Medical Internet Research and other JMIR Publications journals has steadily increased. Editors and peer reviewers involved in the review process for such manuscripts often go through multiple review cycles to enhance the quality and completeness of reporting. The use of reporting guidelines or checklists can help ensure consistency in the quality of submitted (and published) scientific manuscripts and, for example, avoid instances of missing information. In this Editorial, the editors of JMIR Publications journals discuss the general JMIR Publications policy regarding authors’ application of reporting guidelines and specifically focus on the reporting of ML studies in JMIR Publications journals, using the Consolidated Reporting of Machine Learning Studies (CREMLS) guidelines, with an example of how authors and other journals could use the CREMLS checklist to ensure transparency and rigor in reporting.


  • Open Access

    Consolidated Reporting Guidelines for Prognostic and Diagnostic Machine Learning Models (CREMLS) (Preprint)

    by Khaled El Emam, Tiffany I Leung, Bradley Malin, William Klement, Gunther Eysenbach

    UNSTRUCTURED The number of papers presenting machine learning (ML) models that are being submitted to and published in the Journal of Medical Internet Research and other JMIR Publications journals has steadily increased. Editors and peer reviewers involved in the review process for such manuscripts often go through multiple review cycles to enhance the quality and completeness of reporting. The use of reporting guidelines or checklists can help ensure consistency in the quality of submitted (and published) scientific manuscripts and, for example, avoid instances of missing information. In this Editorial, the editors of JMIR Publications journals discuss the general JMIR Publications policy regarding authors’ application of reporting guidelines and specifically focus on the reporting of ML studies in JMIR Publications journals, using the Consolidated Reporting of Machine Learning Studies (CREMLS) guidelines, with an example of how authors and other journals could use the CREMLS checklist to ensure transparency and rigor in reporting.

  • Open Access

    Consolidated Reporting Guidelines for Prognostic and Diagnostic Machine Learning Modeling Studies: Development and Validation

    by William Klement, Khaled El Emam

    Background: The reporting of machine learning (ML) prognostic and diagnostic modeling studies is often inadequate, making it difficult to understand and replicate such studies. To address this issue, multiple consensus and expert reporting guidelines for ML studies have been published. However, these guidelines cover different parts of the analytics lifecycle, and individually, none of them provide a complete set of reporting requirements. Objective: We aimed to consolidate the ML reporting guidelines and checklists in the literature to provide reporting items for prognostic and diagnostic ML in in-silico and shadow mode studies. Methods: We conducted a literature search that identified 192 unique peer-reviewed English articles that provide guidance and checklists for reporting ML studies. The articles were screened by their title and abstract against a set of 9 inclusion and exclusion criteria. Articles that were filtered through had their quality evaluated by 2 raters using a 9-point checklist constructed from guideline development good practices. The average κ was 0.71 across all quality criteria. The resulting 17 high-quality source papers were defined as having a quality score equal to or higher than the median. The reporting items in these 17 articles were consolidated and screened against a set of 6 inclusion and exclusion criteria. The resulting reporting items were sent to an external group of 11 ML experts for review and updated accordingly. The updated checklist was used to assess the reporting in 6 recent modeling papers in JMIR AI. Feedback from the external review and initial validation efforts was used to improve the reporting items. Results: In total, 37 reporting items were identified and grouped into 5 categories based on the stage of the ML project: defining the study details, defining and collecting the data, modeling methodology, model evaluation, and explainability. None of the 17 source articles covered all the reporting items. The study details and data description reporting items were the most common in the source literature, with explainability and methodology guidance (ie, data preparation and model training) having the least coverage. For instance, a median of 75% of the data description reporting items appeared in each of the 17 high-quality source guidelines, but only a median of 33% of the data explainability reporting items appeared. The highest-quality source articles tended to have more items on reporting study details. Other categories of reporting items were not related to the source article quality. We converted the reporting items into a checklist to support more complete reporting. Conclusions: Our findings supported the need for a set of consolidated reporting items, given that existing high-quality guidelines and checklists do not individually provide complete coverage. The consolidated set of reporting items is expected to improve the quality and reproducibility of ML modeling studies.


  • Open Access

    Consolidated Reporting Guidelines for Prognostic and Diagnostic Machine Learning Modeling Studies: Development and Validation (Preprint)

    by William Klement, Khaled El Emam

    BACKGROUND The reporting of machine learning (ML) prognostic and diagnostic modeling studies is often inadequate, making it difficult to understand and replicate such studies. To address this issue, multiple consensus and expert reporting guidelines for ML studies have been published. However, these guidelines cover different parts of the analytics lifecycle, and individually, none of them provide a complete set of reporting requirements. OBJECTIVE We aimed to consolidate the ML reporting guidelines and checklists in the literature to provide reporting items for prognostic and diagnostic ML in in-silico and shadow mode studies. METHODS We conducted a literature search that identified 192 unique peer-reviewed English articles that provide guidance and checklists for reporting ML studies. The articles were screened by their title and abstract against a set of 9 inclusion and exclusion criteria. Articles that were filtered through had their quality evaluated by 2 raters using a 9-point checklist constructed from guideline development good practices. The average κ was 0.71 across all quality criteria. The resulting 17 high-quality source papers were defined as having a quality score equal to or higher than the median. The reporting items in these 17 articles were consolidated and screened against a set of 6 inclusion and exclusion criteria. The resulting reporting items were sent to an external group of 11 ML experts for review and updated accordingly. The updated checklist was used to assess the reporting in 6 recent modeling papers in JMIR AI. Feedback from the external review and initial validation efforts was used to improve the reporting items. RESULTS In total, 37 reporting items were identified and grouped into 5 categories based on the stage of the ML project: defining the study details, defining and collecting the data, modeling methodology, model evaluation, and explainability. None of the 17 source articles covered all the reporting items. The study details and data description reporting items were the most common in the source literature, with explainability and methodology guidance (ie, data preparation and model training) having the least coverage. For instance, a median of 75% of the data description reporting items appeared in each of the 17 high-quality source guidelines, but only a median of 33% of the data explainability reporting items appeared. The highest-quality source articles tended to have more items on reporting study details. Other categories of reporting items were not related to the source article quality. We converted the reporting items into a checklist to support more complete reporting. CONCLUSIONS Our findings supported the need for a set of consolidated reporting items, given that existing high-quality guidelines and checklists do not individually provide complete coverage. The consolidated set of reporting items is expected to improve the quality and reproducibility of ML modeling studies.

  • Open Access

    Reporting and Methodological Observations on Prognostic and Diagnostic Machine Learning Studies

    by Khaled El Emam, William Klement, Bradley Malin

    Common reporting and methodological patterns were observed from the peer reviews of prognostic and diagnostic machine learning modeling studies submitted to JMIR AI. In this editorial, we summarized some key observations to inform future studies and their reporting.


  • Open Access

    Reporting and Methodological Observations on Prognostic and Diagnostic Machine Learning Studies (Preprint)

    by Khaled El Emam, William Klement, Bradley Malin

    UNSTRUCTURED Common reporting and methodological patterns were observed from the peer reviews of prognostic and diagnostic machine learning modeling studies submitted to JMIR AI. In this editorial, we summarized some key observations to inform future studies and their reporting.

  • The validation of chest tube management after lung resection surgery using a random forest classifier

    by William Klement, Sebastien Gilbert, Virginia F. Resende, Donna E. Maziak, Andrew J. E. Seely, Farid M. Shamji, Sudhir R. Sundaresan, Patrick J. Villeneuve, Nathalie Japkowicz


  • Open Access

    A machine learning approach predicts future risk to suicidal ideation from social media data

    by Arunima Roy, Katerina Nikolitch, Rachel McGinn, Safiya Jinah, William Klement, Zachary A. Kaminsky

    Machine learning analysis of social media data represents a promising way to capture longitudinal environmental influences contributing to individual risk for suicidal thoughts and behaviors. Our objective was to generate an algorithm termed “Suicide Artificial Intelligence Prediction Heuristic (SAIPH)” capable of predicting future risk to suicidal thought by analyzing publicly available Twitter data. We trained a series of neural networks on Twitter data queried against suicide associated psychological constructs including burden, stress, loneliness, hopelessness, insomnia, depression, and anxiety. Using 512,526 tweets from N = 283 suicidal ideation (SI) cases and 3,518,494 tweets from 2655 controls, we then trained a random forest model using neural network outputs to predict binary SI status. The model predicted N = 830 SI events derived from an independent set of 277 suicidal ideators relative to N = 3159 control events in all non-SI individuals with an AUC of 0.88 (95% CI 0.86–0.90). Using an alternative approach, our model generates temporal prediction of risk such that peak occurrences above an individual specific threshold denote a ~7 fold increased risk for SI within the following 10 days (OR = 6.7 ± 1.1, P = 9 × 10−71). We validated our model using regionally obtained Twitter data and observed significant associations of algorithm SI scores with county-wide suicide death rates across 16 days in August and in October, 2019, most significantly in younger individuals. Algorithmic approaches like SAIPH have the potential to identify individual future SI risk and could be easily adapted as clinical decision tools aiding suicide screening and risk monitoring using available technologies.