a2z-Abdo-Triage

{0} **FDA** U.S. FOOD & DRUG ADMINISTRATION September 16, 2026 A2z Radiology Ai, Inc. Samir Rajpurkar Chief Executive Officer 292 Newbury St. Unit 235 Boston, Massachusetts 02115 Re: K260906 Trade/Device Name: a2z-Abdo-Triage Regulation Number: 21 CFR 892.2080 Regulation Name: Radiological computer aided triage and notification software Regulatory Class: Class II Product Code: QAS, QFM Dated: August 13, 2026 Received: August 14, 2026 Dear Samir Rajpurkar: We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading. If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register. U.S. Food & Drug Administration 10903 New Hampshire Avenue Silver Spring, MD 20993 www.fda.gov {1} K260906 - Samir Rajpurkar Page 2 FDA's substantial equivalence determination also included the review and clearance of your Predetermined Change Control Plan (PCCP). Under section 515C(b)(1) of the Act, a new premarket notification is not required for a change to a device cleared under section 510(k) of the Act, if such change is consistent with an established PCCP granted pursuant to section 515C(b)(2) of the Act. Under 21 CFR 807.81(a)(3), a new premarket notification is required if there is a major change or modification in the intended use of a device, or if there is a change or modification in a device that could significantly affect the safety or effectiveness of the device, e.g., a significant change or modification in design, material, chemical composition, energy source, or manufacturing process. Accordingly, if deviations from the established PCCP result in a major change or modification in the intended use of the device, or result in a change or modification in the device that could significantly affect the safety or effectiveness of the device, then a new premarket notification would be required consistent with section 515C(b)(1) of the Act and 21 CFR 807.81(a)(3). Failure to submit such a premarket submission would constitute adulteration and misbranding under sections 501(f)(1)(B) and 502(o) of the Act, respectively. Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device" (https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download). Your device is also subject to, among other requirements, the Quality Management System Regulation (QMSR) (21 CFR Part 820), which includes, but is not limited to, ISO 13485 clause 7.3 (Design controls), ISO 13485 clause 8.3 (Nonconforming product), ISO 13485 clause 8.5.2 (Corrective action), and ISO 13485 clause 8.5.3 (Preventative action). Please note that regardless of whether a change requires premarket review, the QMSR requires device manufacturers to review and approve changes to device design and production (ISO 13485 clause 7.3 and ISO 13485 clause 7.5) and document changes and approvals in the Medical Device File (ISO 13485 clause 4.2.3). Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting (reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reporting-combination-products); good manufacturing practice requirements as set forth in the Quality Management System Regulation (QMSR) (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050. All medical devices, including Class I and unclassified devices and combination product device constituent parts are required to be in compliance with the final Unique Device Identification System rule ("UDI Rule"). The UDI Rule requires, among other things, that a device bear a unique device identifier (UDI) on its label and package (21 CFR 801.20(a)) unless an exception or alternative applies (21 CFR 801.20(b)) and that the dates on the device label be formatted in accordance with 21 CFR 801.18. The UDI Rule (21 CFR 830.300(a) and 830.320(b)) also requires that certain information be submitted to the Global Unique Device Identification Database (GUDID) (21 CFR Part 830 Subpart E). For additional information on these {2} K260906 - Samir Rajpurkar Page 3 requirements, please see the UDI System webpage at https://www.fda.gov/medical-devices/device-advice-comprehensive-regulatory-assistance/unique-device-identification-system-udi-system. Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-devices/medical-device-safety/medical-device-reporting-mdr-how-report-medical-device-problems. For comprehensive regulatory information about medical devices and radiation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medical-devices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-devices/device-advice-comprehensive-regulatory-assistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100). Sincerely, Jessica Lamb, PhD Assistant Director Imaging Software Team DHT8B: Division of Radiological Imaging Devices and Electronic Products OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health Enclosure {3} | Indications for Use | | | | --- | --- | --- | | Please type in the marketing application/submission number, if it is known. This textbox will be left blank for original applications/submissions. | K260906 | ? | | Please provide the device trade name(s). | | ? | | a2z-Abdo-Triage | | | | Please provide your Indications for Use below. | | ? | | a2z-Abdo-Triage is a radiological computer-aided triage and notification software indicated for use in the analysis of abdominopelvic CT images acquired with or without intravenous contrast in adults aged 22 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communicating suspected positive cases of the 10 specified findings:Obstructing ureteral stoneHydroureterRetroperitoneal hematomaLarge volume ascitesAcute appendicitisIntraabdominopelvic abscessHemoperitoneum or complex free fluidSplenic traumatic injuryHepatic traumatic injuryRenal traumatic injuryThese findings are intended to be used together as one device.a2z-Abdo-Triage uses an artificial intelligence algorithm to analyze images and flag cases with detected findings in parallel to the ongoing standard of care image interpretation. The device provides analysis results that enable client systems to generate notifications for cases with suspected findings. These results can include DICOM instance UIDs for key images, which are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.The results of a2z-Abdo-Triage are intended to be used in conjunction with other patient information and based on clinicians' professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care. Its results are not intended to be used on a stand-alone basis for clinical-decision making or otherwise preclude clinical assessment of any disease. | | | | Please select the types of uses. | Prescription Use (21 CFR 801 Subpart D)Over-The-Counter Use (21 CFR 801 Subpart C) | ? | | Please select the age group(s) for which the device(s) is to be used. | Neonates/Newborns (Birth to < 29 days old)Infants (29 days old to < 2 years old)Children (2 years old to < 12 years old)Adolescents (12 years old to < 22 years old)Adults (22 years old and greater) | ? | {4} K260906 a2z RADIOLOGY AI 510(k) Summary # Contents 1. Administrative Information ... 1 2. Subject Device Information ... 1 3. Predicate Device Information 2 4. Device Description 2 5. Indications for Use 3 6. Comparison of Technological Characteristics with the Predicate Device ... 3 7. Performance Data 5 7.1. Software Verification and Validation Testing 5 7.2. Performance Assessment 5 Patient Demographics (N=914 received patient pseudonyms) 6 Study-Level Imaging Characteristics (N=965 studies) 6 Condition Prevalence in Test Cohort 7 QFM Condition Performance 8 QAS Condition Performance 8 8. Predetermined Change Control Plan (PCCP) 9 9. Conclusion 10 # 510(k) Summary # 1. Administrative Information | Submitter Name | a2z Radiology AI Inc. | | --- | --- | | Address | 292 Newbury Street, Unit 235, Boston, MA 02115, USA | | Phone Number | +1 508-293-1822 | | Fax Number | N/A | | Company Representative | Samir Rajpurkar | | Email | support@a2zradiology.ai | | Date Summary Prepared | September 15, 2026 | # 2. Subject Device Information | Trade Name | a2z-Abdo-Triage | | --- | --- | | Subject Device K Number | K260906 | | Common Name | Radiological computer aided triage and notification software | | Product Code | QAS, QFM | | Regulation Number | 892.2080 | | Regulatory Class | Class II | | Review Panel | Radiology | {5} 510(k) Summary a2z RADIOLOGY AI ### 3. Predicate Device Information | Predicate Device Name | a2z-Unified-Triage | | --- | --- | | Predicate Device K Number | K252366 | | Common Name | Radiological computer aided triage and notification software | | Product Code | QAS, QFM | | Regulation Number | 892.2080 | | Regulatory Class | Class II | | Review Panel | Radiology | This predicate has not been subject to a design-related recall. No reference devices were used in this submission. ### 4. Device Description a2z-Abdo-Triage is a radiological computer-aided triage and notification software device. The software consists of an algorithmic component that supports both cloud-based and on-premises deployment on standard server hardware. The device processes abdomen/pelvis CT images from clinical imaging systems, analyzing them using artificial intelligence algorithms to detect suspected cases of 10 abdominopelvic conditions: Obstructing ureteral stone, Hydroureter, Retroperitoneal hematoma, Large volume ascites, Acute appendicitis, Intraabdominopelvic abscess, Hemoperitoneum or complex free fluid, Splenic traumatic injury, Hepatic traumatic injury, and Renal traumatic injury. Following the AI processing, the analysis results are returned to the client system for worklist prioritization. When a suspected case is detected, the software provides analysis results that enable the client system to generate appropriate notifications. These results can include DICOM instance UIDs for key images, which are for informational purposes only, do not contain any marking of the findings, and are not intended for diagnostic use beyond notification. Integration with clinical imaging systems facilitates efficient triage by enabling prioritization of suspect cases for review of the relevant original images in the PACS. Thus, the suspect case receives attention earlier than would have been the case in the standard of care practice alone. Algorithm Architecture: The artificial intelligence algorithms implement deep learning models that analyze CT studies and identify suspected positive findings. For each detected finding, the software provides one or more image references most likely to show the condition; those references are informational only. A finding that is not reported is not a diagnostic negative result. Training Database: The algorithms were developed using a multi-site CT dataset with diverse imaging characteristics. The filed findings were evaluated on the independent abdominopelvic clinical test cohort. Data Independence: No received patient pseudonym appeared in both the development and test datasets, and the institutions contributing the test cohort contributed no studies to the training or tuning splits. {6} 510(k) Summary a2z RADIOLOGY AI ## 5. Indications for Use a2z-Abdo-Triage is a radiological computer-aided triage and notification software indicated for use in the analysis of abdominopelvic CT images acquired with or without intravenous contrast in adults aged 22 and older. The device is intended to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communicating suspected positive cases of the 10 specified findings: - Obstructing ureteral stone - Hydroureter - Retroperitoneal hematoma - Large volume ascites - Acute appendicitis - Intraabdominopelvic abscess - Hemoperitoneum or complex free fluid - Splenic traumatic injury - Hepatic traumatic injury - Renal traumatic injury These findings are intended to be used together as one device. a2z-Abdo-Triage uses an artificial intelligence algorithm to analyze images and flag cases with detected findings in parallel to the ongoing standard of care image interpretation. The device provides analysis results that enable client systems to generate notifications for cases with suspected findings. These results can include DICOM instance UIDs for key images, which are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device. The results of a2z-Abdo-Triage are intended to be used in conjunction with other patient information and based on clinicians' professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care. Its results are not intended to be used on a stand-alone basis for clinical-decision making or otherwise preclude clinical assessment of any disease. ## 6. Comparison of Technological Characteristics with the Predicate Device The subject device, a2z-Abdo-Triage, is substantially equivalent to the predicate device a2z-Unified-Triage (K252366), a cleared radiological computer-aided triage and notification software for the abdominopelvic CT region. The subject and predicate are both radiological computer-aided triage and notification software using artificial intelligence algorithms. Both provide notifications for worklist prioritization. Neither removes cases from standard-care reading queues nor de-prioritizes cases; both operate in parallel with standard care. The technology type, intended use for triage, and principles of operation are shared with the predicate. Both provide positive notifications for worklist prioritization. The subject device reports {7} 510(k) Summary a2z RADIOLOGY AI suspected positive findings; an unreported finding is not a diagnostic negative result. The specified finding sets differ, and each subject-device finding is supported by standalone performance testing. The table below compares key features. Table 1. Comparison with predicate device. | Characteristic | Subject Device: a2z-Abdo-Triage | Predicate Device: a2z-Unified-Triage (K252366) | | --- | --- | --- | | Manufacturer | a2z Radiology AI Inc. | a2z Radiology AI Inc. | | Regulation Number | 892.2080 | 892.2080 | | Regulatory Class | Class II | Class II | | Product Code | QAS, QFM | QAS, QFM | | Regulation Name | Radiological computer aided triage and notification software | Radiological computer aided triage and notification software | | Device Property | SaMD (Software as a Medical Device) | SaMD (Software as a Medical Device) | | Indications for Use | Radiological computer-aided triage and notification software for analysis of abdominopelvic CT images acquired with or without intravenous contrast in adults aged 22+. Assists in workflow triage by flagging 10 specified findings. | Radiological computer-aided triage and notification software for analysis of abdominal/pelvic CT images in adults aged 22+. Assists in workflow triage by flagging 7 abdominopelvic findings. | | Technical Characteristics | | | | Input | Abdominopelvic CT images acquired with or without intravenous contrast | Abdomen/Pelvis CT images (with or without contrast) | | Output | Suspected positive findings for worklist prioritization and informational DICOM instance UIDs for key images; unreported findings are not negative diagnostic results | Notification to image and order management system for worklist prioritization | | Algorithm Type | Non-adaptive machine learning (Deep Learning - Neural Networks) | Non-adaptive machine learning (Deep Learning - Neural Networks) | | Intended Users | Appropriately trained medical specialists qualified to interpret abdominal/pelvic CT studies | Appropriately trained medical specialists qualified to interpret abdominal/pelvic CT studies | | Target Population | Adults (22 years and older) | Adults (22 years and older) | | Location of anatomical structures | Abdomen and Pelvis | Abdomen and Pelvis | | Imaging Modality | Computed Tomography (CT) | Computed Tomography (CT) | | Intended Use Environment | Hospital environment or other clinical settings with DICOM-compliant CT imaging and IT infrastructure | Hospital environment or other clinical settings with DICOM-compliant CT imaging and IT infrastructure | | Performance | Met pre-specified performance targets for all findings | Met pre-specified performance targets for all findings | {8} 510(k) Summary | Characteristic | Subject Device: a2z-Abdo-Triage | Predicate Device: a2z-Unified-Triage (K252366) | | --- | --- | --- | | Turnaround Time | Technical processing interval – mean 133.52 seconds (95% CI: 129.26-137.78), median 97.19 seconds, 95th percentile 276.74 seconds | Filed predicate turnaround values – mean 58.39 seconds (95% CI: 56.11-60.68), median 55.02 seconds, 95th percentile 90.36 seconds. Measurement boundaries differed, so the values are not a direct speed comparison. | | Software device that operates on off-the-shelf hardware | Yes. Software supports both cloud-based and on-premises deployment on standard server hardware | Yes. Interfaces with image and order management systems | Table 1 compares the subject device to the predicate. The Turnaround Time row reproduces each device's own filed turnaround-time figures as measured; the measurement boundaries differ between the two devices, and no direct speed comparison is claimed. ## 7. Performance Data ### 7.1. Software Verification and Validation Testing Software verification and validation testing was conducted. ### 7.2. Performance Assessment A standalone performance assessment was performed for a2z-Abdo-Triage to validate the accuracy of detecting the 10 findings against a reference standard established by U.S. board-certified radiologists. #### 7.2.1. Reader Reference Standard The reference standard for the 989-case test dataset used a 2+1 methodology with U.S. board-certified radiologists. Each study-finding classification received two independent U.S. board-certified radiologist determinations. When those readers disagreed, an independent U.S. board-certified radiologist adjudicator provided the tiebreaking determination. Readers were blinded to device output. #### 7.2.2. Development and Test Dataset a2z-Abdo-Triage was evaluated on a held-out reader-reference-standard cohort of 989 abdominopelvic CT studies. After prespecified exclusions, 965 CT studies were included in the performance analysis. The enriched test design provided positive representation for each filed finding; the cohort's observed demographics and imaging characteristics are reported below. #### 7.2.3. Test Cohort Characteristics The 965 studies in the analytic cohort are associated with 914 received patient pseudonyms. Some pseudonyms are associated with more than one study. Age and sex are reported at the received- {9} a2z RADIOLOGY AI # 510(k) Summary pseudonym level; imaging characteristics (site, CT scanner manufacturer, contrast status, slice thickness) are reported at the study level because acquisition parameters can vary across studies. # Patient Demographics (N=914 received patient pseudonyms) # Age Distribution - Mean: 54.6 years (SD: 19.7) - Median: 54 years - Range: 22-90 years | Age Group | N | Percentage | | --- | --- | --- | | 22-39 | 262 | 28.7% | | 40-59 | 272 | 29.8% | | 60-79 | 260 | 28.4% | | 80+ | 120 | 13.1% | Sex Distribution | Sex | N | Percentage | | --- | --- | --- | | Female | 435 | 47.6% | | Male | 477 | 52.2% | | Not available | 2 | 0.2% | # Study-Level Imaging Characteristics (N=965 studies) Site Distribution | Site | N | Percentage | | --- | --- | --- | | Site 1 | 548 | 56.8% | | Site 2 | 294 | 30.5% | | Site 3 | 101 | 10.5% | | Not available | 22 | 2.3% | Geographic Distribution | State | N | Percentage | | --- | --- | --- | | FL | 548 | 56.8% | | TX | 294 | 30.5% | | OH | 101 | 10.5% | | Not available | 22 | 2.3% | CT Scanner Manufacturer Distribution | Manufacturer | N | Percentage | | --- | --- | --- | | SIEMENS | 497 | 51.5% | | GE | 391 | 40.5% | {10} a2z RADIOLOGY AI # 510(k) Summary | Manufacturer | N | Percentage | | --- | --- | --- | | Canon | 34 | 3.5% | | TOSHIBA | 34 | 3.5% | | FUJI | 7 | 0.7% | | Other | 2 | 0.2% | # Contrast Status | Contrast | N | Percentage | | --- | --- | --- | | With contrast | 558 | 57.8% | | Without contrast | 407 | 42.2% | # Slice Thickness Distribution - Range: 0.6-5.0mm, Mean: 2.9mm, Median: 2.0mm | Slice Thickness | N | Percentage | | --- | --- | --- | | <5mm | 707 | 73.3% | | 5mm | 258 | 26.7% | ### 7.2.4. Clinical Subgroups and Confounders The test dataset included positive representation for each of the 10 target findings and cases with concurrent findings. Descriptive stratified results were evaluated for the observed demographic, imaging, site, and disease-presentation groups. Sparse or non-estimable strata were identified; no separate per-stratum acceptance criterion is claimed. ### 7.2.5. Primary Endpoints (Prevalence) Condition Prevalence in Test Cohort | Condition | N Total | N Positive | Prevalence | | --- | --- | --- | --- | | Acute appendicitis | 965 | 76 | 7.9% | | Hemoperitoneum or complex free fluid | 965 | 172 | 17.8% | | Hepatic traumatic injury | 965 | 59 | 6.1% | | Hydroureter | 965 | 162 | 16.8% | | Intraabdominopelvic abscess | 965 | 109 | 11.3% | | Large volume ascites | 965 | 79 | 8.2% | | Obstructing ureteral stone | 965 | 118 | 12.2% | | Renal traumatic injury | 965 | 70 | 7.3% | | Retroperitoneal hematoma | 965 | 116 | 12.0% | | Splenic traumatic injury | 965 | 72 | 7.5% | {11} 510(k) Summary a2z RADIOLOGY AI #### 7.2.6. QFM Conditions QFM Condition Performance | Condition | Available Operating Points and Performance | | --- | --- | | Hydroureter (QFM)Total N=965,Positive=162AUC: 0.984[0.972-0.996] | High Sensitivity: Se 97.5% [93.8-99.0%], Sp 88.4% [86.0-90.5%]Sensitivity Biased: Se 94.4% [89.8-97.0%], Sp 94.4% [92.6-95.8%]Balanced: Se 91.4% [86.0-94.8%], Sp 96.8% [95.3-97.8%] | | Large volume ascites (QFM)Total N=965,Positive=79AUC: 0.995[0.991-0.998] | High Specificity: Se 100.0% [95.4-100.0%], Sp 83.1% [80.5-85.4%] | | Obstructing ureteral stone (QFM)Total N=965,Positive=118AUC: 0.994[0.989-0.998] | High Sensitivity: Se 99.2% [95.4-99.9%], Sp 85.4% [82.8-87.6%]Sensitivity Biased: Se 97.5% [92.8-99.1%], Sp 95.3% [93.6-96.5%]Balanced: Se 94.1% [88.3-97.1%], Sp 98.8% [97.8-99.4%] | #### 7.2.7. QAS Conditions QAS Condition Performance | Condition | Available Operating Points and Performance | | --- | --- | | Acute appendicitis (QAS)Total N=965,Positive=76 | High Sensitivity: Se 97.4% [90.9-99.3%], Sp 86.6% [84.2-88.7%]Sensitivity Biased: Se 94.7% [87.2-97.9%], Sp 96.6% [95.2-97.6%]Balanced: Se 88.2% [79.0-93.6%], Sp 99.1% [98.2-99.5%] | | Hemoperitoneum or complex free fluid (QAS)Total N=965,Positive=172 | High Specificity: Se 86.6% [80.7-90.9%], Sp 83.1% [80.3-85.5%] | | Hepatic traumatic injury (QAS)Total N=965,Positive=59 | Balanced: Se 91.5% [81.6-96.3%], Sp 89.5% [87.3-91.3%]High Specificity: Se 81.4% [69.6-89.3%], Sp 96.0% [94.5-97.1%] | | Intraabdominopelvic abscess (QAS)Total N=965,Positive=109 | Sensitivity Biased: Se 90.8% [83.9-94.9%], Sp 87.5% [85.1-89.5%] | | Renal traumatic injury (QAS)Total N=965,Positive=70 | Sensitivity Biased: Se 95.7% [88.1-98.5%], Sp 85.1% [82.7-87.3%]Balanced: Se 94.3% [86.2-97.8%], Sp 95.3% [93.7-96.5%] | {12} a2z RADIOLOGY AI 510(k) Summary | Condition | Available Operating Points and Performance | | --- | --- | | Retroperitoneal hematoma (QAS) Total N=965, Positive=116 | Balanced: Se 87.9% [80.8-92.7%], Sp 92.3% [90.4-93.9%] | | Splenic traumatic injury (QAS) Total N=965, Positive=72 | Sensitivity Biased: Se 97.2% [90.4-99.2%], Sp 85.2% [82.7-87.4%] Balanced: Se 88.9% [79.6-94.3%], Sp 95.0% [93.3-96.2%] | ### 7.2.8. Performance Discussion The standalone performance assessment validated 10 abdominopelvic findings comprising 3 QFM findings (Hydroureter, Large volume ascites, Obstructing ureteral stone) and 7 QAS findings (Acute appendicitis, Hemoperitoneum or complex free fluid, Hepatic traumatic injury, Intraabdominopelvic abscess, Renal traumatic injury, Retroperitoneal hematoma, Splenic traumatic injury). All 10 findings met pre-specified acceptance criteria, consistent with the statistical methodology used in the predicate device submission. QFM AUC values ranged from 0.984 to 0.995, and QAS AUC values ranged from 0.945 to 0.985. Each finding met its prespecified, finding-type-specific acceptance criteria at one or more reported operating points. Stratified subgroup analyses were conducted across patient demographics, clinical sites and regions, imaging acquisition parameters, and concurrent findings. These analyses are descriptive and supportive: estimates and confidence intervals characterize the observed data, but no formal per-stratum acceptance criterion or claim of absence of subgroup degradation is applied. Sparse or non-estimable strata were identified. The technical processing interval had a mean of 133.52 seconds, a median of 97.19 seconds, and a 95th percentile of 276.74 seconds per study. This interval excludes deployment-specific overhead and does not represent end-to-end clinical-workflow turnaround. In summary, a2z-Abdo-Triage met the prespecified performance criteria for the 10 validated findings. These results support the substantial-equivalence determination for the device's intended triage-and-notification use. ### 8. Predetermined Change Control Plan (PCCP) This submission contains a Predetermined Change Control Plan (PCCP). The PCCP does not authorize adaptive algorithms that continuously learn in the field. A modified model is trained, validated, and locked before release. Labeling is updated for implemented PCCP changes as specified in the plan. The PCCP specifies possible modifications to a2z-Abdo-Triage and the verification, validation, risk-management, and acceptance criteria that must be satisfied before an in-scope modification is implemented. The PCCP applies to the 10 conditions validated in this 510(k) submission: Obstructing ureteral stone, Hydroureter, Retroperitoneal hematoma, Large volume ascites, Acute {13} 510(k) Summary a2z RADIOLOGY AI appendicitis, Intraabdominopelvic abscess, Hemoperitoneum or complex free fluid, Splenic traumatic injury, Hepatic traumatic injury, and Renal traumatic injury. For each condition, the PCCP covers operating points that meet the pre-specified performance criteria upon revalidation. The planned modifications include: - Training data volume and diversity expansion - Training data quality and annotation refinement - Ensemble composition optimization • Data augmentation parameter adjustments - Training parameter optimization • Weight initialization method selection - Neural network architecture component modifications - Input data preprocessing adjustments The PCCP defines bounded specifications and prespecified validation requirements. Each modification must complete the applicable validation requirements before implementation. For substantive modifications, these requirements include independent test data, U.S. board-certified-radiologist reference standards, data sequestration, descriptive subgroup analysis, and prespecified statistical acceptance criteria. The modification protocol incorporates impact assessment considerations and specifies requirements for data management, including data sources, collection, storage, and sequestration, as well as documentation and data re-use practices. Detailed validation activities, testing methodologies, and performance requirements have been established for each modification. Each change must complete the specified verification and validation before implementation and meet the PCCP acceptance criteria. ## 9. Conclusion The subject device, a2z-Abdo-Triage, and the predicate, a2z-Unified-Triage, are substantially equivalent. Both are software devices intended to aid in the prioritization and triage of radiological images using AI algorithms. They share the same fundamental scientific technology and principles of operation. While the specific clinical findings differ, the intended use for triage, the AI-based approach, the integration into clinical workflows, and the parallel nature of operation are highly similar. Both devices target the same anatomical region (abdomen/pelvis) and aim to reduce turnaround time through preemptive triage without altering the standard of care. The performance data show that a2z-Abdo-Triage met the prespecified criteria for its intended triage-and-notification use and support the determination that a2z-Abdo-Triage is substantially equivalent to the predicate device.