BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:US/Eastern BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260928T141647Z UID:8D78B2BE-5911-42FE-A11E-93BC12A2290F DTSTART;TZID=US/Eastern:20261020T190000 DTEND;TZID=US/Eastern:20261020T200000 DESCRIPTION:From Pixels to Prognosis:\nAI\, Imaging\, and Alternative Physi ological Signals for Chronic Obstructive Pulmonary Disease\n\nPart I · AI and Quantitative Imaging for COPD Chronic obstructive pulmonary disease ( COPD) is the third leading cause of death worldwide and remains underdiagn osed in more than half of affected individuals. Spirometry\, the current d iagnostic standard\, is insensitive to early structural change\, and quant itative information present in every chest CT scan is largely unexploited in routine radiology practice. This talk reviews a decade of work developi ng artificial intelligence and quantitative imaging methods that recover d isease-relevant structural\, mechanical\, and textural information invisib le to the human eye\, drawing on validation in the COPDGene\, SPIROMICS\, and National Lung Screening Trial cohorts.\n\nThe talk covers three comple mentary computational threads: (i) geometry-derived CT biomarkers includin g airway fractal dimension\, tapering\, surface-area-to-volume ratio\, and CT-derived dysanapsis\, all independently associated with respiratory mor bidity and mortality\; (ii) silent zones of lung disease\, where deep lear ning applied to visually normal parenchyma reveals subclinical radiomic an d mechanical signatures that predict emphysema onset\, exacerbation\, and mortality years before conventional criteria trigger\; and (iii) deep neur al network analysis of raw spirometry flow-volume curves\, which recovers CT-defined structural phenotype from a single breath and separates true ve ntilatory restriction from obstruction. The talk closes with translation o f these methods through UAB spin-offs and where the field is moving with v ision-language foundation models for multimodal chest imaging at the point of care.\n\nPart II · Deep Learning on the ECG for COPD Detection This t alk reviews recent evidence that convolutional neural networks can identif y COPD directly from the 12-lead electrocardiogram (ECG)\, drawing on the study by Vaid et al. Because COPD produces measurable cardiopulmonary effe cts\, the ECG\, a ubiquitous\, low-cost\, and effort-independent signal\, may carry latent information about airflow obstruction. The Mount Sinai CN N model was trained on 208\,231 ECGs from 18\,225 COPD cases matched to 49 \,356 controls and evaluated on an internal holdout\, an external hospital cohort\, and the UK Biobank\, achieving AUC of 0.80\, 0.82\, and 0.75 res pectively. The talk covers the model architecture\, its linkage to concurr ent spirometry\, the explainability analyses identifying P-wave changes as the most informative ECG feature for COPD\, and how ECG-based prediction could complement the imaging pipelines described in Part I and extend COPD screening into resource-limited settings.\n\nSpeaker(s): Kelly Powell\, M BA \, Sandeep Bodduluri\, MS\, PhD\n\nAgenda: \nWEBINAR: 7:00 - 8:00 P.M.\ n\nThe Zoom Webinar link and password will be forwarded to all registered participants after Noon on the day of the meeting. Check your spam folder if you don't see the email.\n\nWebinar is open to all.\n\nPDH certificates are available and an evaluation form will be emailed to you after the mee ting. PDH certificate are sent by IEEE USA 3-4 weeks after the meeting.\n\ nVirtual: https://events.vtools.ieee.org/m/579663 LOCATION:Virtual: https://events.vtools.ieee.org/m/579663 ORGANIZER:ieee@gpamg.org SEQUENCE:58 SUMMARY:From Pixels to Prognosis: AI\, Imaging\, and Alternative Physiologi cal Signals for Chronic Obstructive Pulmonary Disease URL;VALUE=URI:https://events.vtools.ieee.org/m/579663 X-ALT-DESC:Description: <br /><p><strong><span style="mso-bidi-font-weight: bold\;">From Pixels to Prognosis:<br>AI\, Imaging\, and Alternative Physi ological Signals for Chronic Obstructive Pulmonary Disease<br></span></str ong></p>\n<p class="MsoNormal" style="margin: 11.0pt 0in 4.0pt 0in\;"><str ong><span style="font-size: 13.0pt\;">Part I<span style="mso-spacerun: yes \;">&nbsp\; </span>&middot\;<span style="mso-spacerun: yes\;">&nbsp\; </sp an>AI and Quantitative Imaging for COPD<span style="mso-spacerun: yes\;"> </span></span></strong>Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide and remains underdiagnosed in more than half of affected individuals. Spirometry\, the current diagnostic st andard\, is insensitive to early structural change\, and quantitative info rmation present in every chest CT scan is largely unexploited in routine r adiology practice. This talk reviews a decade of work developing artificia l intelligence and quantitative imaging methods that recover disease-relev ant structural\, mechanical\, and textural information invisible to the hu man eye\, drawing on validation in the COPDGene\, SPIROMICS\, and National Lung Screening Trial cohorts.</p>\n<p class="MsoNormal" style="margin-bot tom: 6.0pt\; line-height: 120%\;">The talk covers three complementary comp utational threads: (i) geometry-derived CT biomarkers including airway fra ctal dimension\, tapering\, surface-area-to-volume ratio\, and CT-derived dysanapsis\, all independently associated with respiratory morbidity and m ortality\; (ii) <em>silent zones of lung disease</em>\, where deep learnin g applied to visually normal parenchyma reveals subclinical radiomic and m echanical signatures that predict emphysema onset\, exacerbation\, and mor tality years before conventional criteria trigger\; and (iii) deep neural network analysis of raw spirometry flow-volume curves\, which recovers CT- defined structural phenotype from a single breath and separates true venti latory restriction from obstruction. The talk closes with translation of t hese methods through UAB spin-offs and where the field is moving with visi on-language foundation models for multimodal chest imaging at the point of care.</p>\n<p class="MsoNormal" style="margin: 11.0pt 0in 4.0pt 0in\;"><s trong><span style="font-size: 13.0pt\;">Part II<span style="mso-spacerun: yes\;">&nbsp\; </span>&middot\;<span style="mso-spacerun: yes\;">&nbsp\; < /span>Deep Learning on the ECG for COPD Detection<span style="mso-spacerun : yes\;"> &nbsp\;</span></span></strong>This talk reviews recent evidence that convolutional neural networks can identify COPD directly from the 12- lead electrocardiogram (ECG)\, drawing on the study by Vaid et al. Because COPD produces measurable cardiopulmonary effects\, the ECG\, a ubiquitous \, low-cost\, and effort-independent signal\, may carry latent information about airflow obstruction. The Mount Sinai CNN model was trained on 208\, 231 ECGs from 18\,225 COPD cases matched to 49\,356 controls and evaluated on an internal holdout\, an external hospital cohort\, and the UK Biobank \, achieving AUC of 0.80\, 0.82\, and 0.75 respectively. The talk covers t he model architecture\, its linkage to concurrent spirometry\, the explain ability analyses identifying P-wave changes as the most informative ECG fe ature for COPD\, and how ECG-based prediction could complement the imaging pipelines described in Part I and extend COPD screening into resource-lim ited settings.</p><br /><br />Agenda: <br /><p><strong>WEBINAR:</strong>&n bsp\;7:00 - 8:00 P.M.&nbsp\;</p>\n<p>The Zoom Webinar link and password wi ll be forwarded to all registered participants after Noon on the day of th e meeting. <strong>Check your spam folder if you don't see the email.&nbsp \;</strong></p>\n<p>Webinar is open to all.</p>\n<p>PDH certificates are a vailable and an evaluation form will be emailed to you after the meeting. PDH certificate are sent by IEEE USA 3-4 weeks after the meeting.</p> END:VEVENT END:VCALENDAR