Background Artificial intelligence (AI) systems performing at radiologist-like levels in the evaluation of digital mammography (DM) would improve breast cancer screening accuracy and efficiency. We aimed to compare the stand-alone performance of an AI system to that of radiologists in detecting breast cancer in DM. Methods Nine multi-reader, multi-case study datasets previously used for different research purposes in seven countries were collected. Each dataset consisted of DM exams acquired with systems from four different vendors, multiple radiologists’ assessments per exam, and ground truth verified by histopathological analysis or follow-up, yielding a total of 2652 exams (653 malignant) and interpretations by 101 radiologists (28 296 independent interpretations). An AI system analyzed these exams yielding a level of suspicion of cancer present between 1 and 10. The detection performance between the radiologists and the AI system was compared using a noninferiority null hypothesis at a margin of 0.05. Results The performance of the AI system was statistically noninferior to that of the average of the 101 radiologists. The AI system had a 0.840 (95% confidence interval [CI] = 0.820 to 0.860) area under the ROC curve and the average of the radiologists was 0.814 (95% CI = 0.787 to 0.841) (difference 95% CI = −0.003 to 0.055). The AI system had an AUC higher than 61.4% of the radiologists. Conclusions The evaluated AI system achieved a cancer detection accuracy comparable to an average breast radiologist in this retrospective setting. Although promising, the performance and impact of such a system in a screening setting needs further investigation.

Clinical performance of tomosynthesis in one view at the same total dose as standard screen-film mammography is not inferior to digital mammography in two views.

Purpose To study the feasibility of automatically identifying normal digital mammography (DM) exams with artificial intelligence (AI) to reduce the breast cancer screening reading workload. Methods and materials A total of 2652 DM exams (653 cancer) and interpretations by 101 radiologists were gathered from nine previously performed multi-reader multi-case receiver operating characteristic (MRMC ROC) studies. An AI system was used to obtain a score between 1 and 10 for each exam, representing the likelihood of cancer present. Using all AI scores between 1 and 9 as possible thresholds, the exams were divided into groups of low- and high likelihood of cancer present. It was assumed that, under the pre-selection scenario, only the high-likelihood group would be read by radiologists, while all low-likelihood exams would be reported as normal. The area under the reader-averaged ROC curve (AUC) was calculated for the original evaluations and for the pre-selection scenarios and compared using a non-inferiority hypothesis. Results Setting the low/high-likelihood threshold at an AI score of 5 (high likelihood > 5) results in a trade-off of approximately halving (− 47%) the workload to be read by radiologists while excluding 7% of true-positive exams. Using an AI score of 2 as threshold yields a workload reduction of 17% while only excluding 1% of true-positive exams. Pre-selection did not change the average AUC of radiologists (inferior 95% CI > − 0.05) for any threshold except at the extreme AI score of 9. Conclusion It is possible to automatically pre-select exams using AI to significantly reduce the breast cancer screening reading workload. Key Points • There is potential to use artificial intelligence to automatically reduce the breast cancer screening reading workload by excluding exams with a low likelihood of cancer. • The exclusion of exams with the lowest likelihood of cancer in screening might not change radiologists’ breast cancer detection performance. • When excluding exams with the lowest likelihood of cancer, the decrease in true-positive recalls would be balanced by a simultaneous reduction in false-positive recalls.

BACKGROUND The feasibility and efficacy of concomitant chemotherapy and highly active antiretroviral therapy (HAART) is still unknown in patients with human immunodeficiency virus (HIV)‐related malignancies. To evaluate the impact of chemotherapy plus HAART on the clinical course of patients with HIV‐related, systemic, non‐Hodgkin lymphoma (HIV‐NHL), the authors compared retrospectively a group of 24 patients with HIV‐NHL who were treated with the cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy regimen plus HAART with a group of 80 patients who were treated with CHOP chemotherapy or a CHOP‐like regimen (i.e., cyclophosphamide, doxorubicin, teniposide, and prednisone with vincristine plus bleomycin) without receiving antiretroviral therapy. METHODS All patients were enrolled in two sequential trials performed at the Aviano Cancer Center, Italy, from April 1988 to December 1998. HAART was included with combination therapy from January 1997. Antiretroviral regimens consisted of two reverse transcriptase inhibitors and one protease inhibitor. RESULTS The two treatment groups were well matched with regard to patient demographics, NHL characteristics, HIV status, and treatment, i.e., the number of cycles and chemotherapy dose. The response rates were similar between the two groups. Severe anemia (Grade 3–4 according to the World Health Organization criteria) was significantly greater in the patients who received CHOP‐HAART compared with the patients who received CHOP alone (33% vs. 7%, respectively; P = 0.001). Leukopenia was similar between the two groups, but colony stimulating factor support was significantly greater in the CHOP‐HAART group than in the control group (92% vs. 66%, respectively; P = 0.03). Seventeen percent of CHOP‐HAART patients developed severe autonomic neurotoxicity, whereas none of the CHOP patients developed neurotoxicity (P = 0.002). At similar median follow‐up, opportunistic infection (OI) rates and mortality were significantly lower in the CHOP‐HAART patients than in the CHOP patients (18% vs. 52%, respectively; P = 0.05; and 38% vs. 85%, respectively; P = 0.001). The median survival for CHOP‐HAART patients was not reached, whereas the medial survival of CHOP patients was 7 months (P = 0.03). CONCLUSIONS The combination of CHOP plus HAART is feasible and may reduce the morbidity from OIs in HIV‐NHL patients. However, careful attention must be directed to cross toxicity and possible pharmacokinetic interactions between antiretroviral and antineoplastic drugs. The impact of the combined chemotherapy plus HAART treatment on patient survival needs urgently to be evaluated in prospective studies. Cancer 2001;91:155–63. © 2001 American Cancer Society.