Article Figures & Data
Figures
- Study Flowchart
Development and real-world deployment of an optical character recognition (OCR)-enabled One-ID digital cervical cancer screening platform (web console + WeChat Mini Program). Phase I comprised system architecture design and integration of OCR to capture and standardize identity information via a unified national identification number (One-ID). Phase II involved implementation across six counties with a descriptive pre-post evaluation. The end-to-end digital workflow included community outreach and appointment booking, on-site registration with eligibility verification, specimen collection, laboratory processing and result release, automated recall, colposcopy attendance, biopsy and histopathology, treatment, digital follow-up, and management completion. Over-screening (screening at intervals shorter than guideline recommendations) decreased from 12.64% to 0.17% (-12.47 percentage points; -98.7%), follow-up adherence and completion of CIN2+ management improved, and CIN2+ detection rate increased in parallel with broader adoption of HPV testing. These patterns were consistent across age groups, counties, and service delivery strategies. (This pre-post assessment was descriptive and not designed to infer causal effects). CIN2+, cervical intraepithelial neoplasia grade 2 or worse; HPV, human papillomavirus; OCR, optical character recognition; One-ID, unified national identity number; pp, percentage points; QC, quality control.
- Figure 1
Participant flow and phased deployment of the digital platform across six study counties in China. (A) Participant flow for all screening encounters included in the analysis (2021–2024, combined). Boxes show the number of women at each step—from registration → primary screening → result notification → recall → colposcopy → biopsy/histopathology → CIN2+ diagnosis → treatment/management—with losses to follow-up and key exclusions displayed on the right (e.g., outside the 35–64 age range, missing ID number, lost test result, low-quality screening). (B) Annual counts by county and study phase. The “No-digital phase” (2021–2023) reflects paper-based records, and the “Digital phase” (2024: OCR module; 2025: mini-program pilot) reflects real-time electronic capture with One-ID–based deduplication. From 2024 onward, sites predominantly adopted HPV-based primary screening; therefore, between-year CIN2+ comparisons are descriptive. Study counties: Zezhou and Xiangyuan (Shanxi), Yanting and Shimian (Sichuan), and Xinping and Mangshi (Yunnan). CIN2+, cervical intraepithelial neoplasia grade 2 or worse; HPV, human papillomavirus; OCR, optical character recognition; One-ID, unified national identity number.
- Figure 2
Digital platform implementation and age-specific follow-up rates. (A) Age-specific colposcopy follow-up completion rate before versus after platform roll-out, pooled across the six study counties. “Non-digital workflow” indicates paper-based management (pre-implementation) and “Digital workflow” indicates the OCR-enabled One-ID system (post-implementation). Points are the colposcopy completion proportion among women referred in the index year. Shaded ribbons show 95% binomial CIs. (B) Annual age-specific follow-up rates by study phase: 2021-Pre; 2022-Pre; 2023-Pre (pre-implementation); 2023-Post; 2024-Post (after mid-2023 roll-out and during full-year 2024 operation, respectively). From 2024 onward, sites predominantly adopted HPV-based primary screening. Thus, between-year differences are descriptive. Overall, follow-up completion rate was higher across all age groups (35–39, 40–44, 45–49, 50–54, 55–59, and 60–64 y) in the digital workflow than in the non-digital workflow. CI, confidence interval; OCR, optical character recognition; One-ID, unified national identity number; Post, post-implementation; Pre, pre-implementation.
Tables
Item 2021 (N, %) 2022 (N, %) 2023 (N, %) 2024 (N, %) Overall (N, %) Overall 31,927 (100%) 35,873 (100%) 46,098 (100%) 40,080 (100%) 153,978 (100%) Age 35–44 10,127 (31.72%) 10,347 (28.84%) 13,376 (29.02%) 10,700 (26.70%) 44,550 (28.93%) 45–54 15,192 (47.58%) 16,768 (46.74%) 20,750 (45.01%) 16,589 (41.39%) 69,299 (45.01%) 55–64 6,608 (20.70%) 8,758 (24.41%) 11,972 (25.97%) 12,791 (31.91%) 40,129 (26.06%) Site Zezhou County 7,239 (22.67%) 4,342 (12.10%) 8,874 (19.25%) 7,058 (17.61%) 27,513 (17.87%) Xiangyuan County 5,179 (16.22%) 5,992 (16.70%) 5,247 (11.38%) 5,646 (14.09%) 22,064 (14.33%) Yanting County 3,917 (12.27%) 6,982 (19.46%) 7,003 (15.19%) 7,621 (19.01%) 25,523 (16.58%) Shimian County 3,556 (11.14%) 4,002 (11.16%) 5,395 (11.70%) 3,180 (7.93%) 16,133 (10.48%) Mangshi 8,272 (25.91%) 10,438 (29.10%) 11,636 (25.24%) 8,069 (20.13%) 38,415 (24.95%) Xinping County 3,764 (11.79%) 4,117 (11.48%) 7,943 (17.23%) 8,506 (21.22%) 24,330 (15.80%) Screening strategy HPV 4,520 (14.16%) 4,838 (13.49%) 3,269 (7.09%) 2,470 (6.16%) 15,097 (9.80%) HPV + cytology 96 (0.30%) 1,110 (3.09%) 11,771 (25.53%) 9,318 (23.25%) 22,295 (14.48%) VIA + cytology triage 8,272 (25.91%) 0 0 0 8,272 (5.37%) Cytology 19,039 (59.63%) 29,925 (83.42%) 10,088 (21.88%) 0 59,052 (38.35%) HPV + VIA/VILI triage 0 0 9,278 (20.13%) 4,667 (11.64%) 13,945 (9.06%) HPV + cytology triage 0 0 11,692 (25.36%) 16,435 (41.01%) 28,127 (18.27%) HPV + cytology + VIA/VILI 0 0 0 7,190 (17.94%) 7,190 (4.67%) Primary screening result Positive 1,778 (5.57%) 2,203 (6.14%) 2,790 (6.05%) 2,775 (6.92%) 9,546 (6.20%) Negative 30,149 (94.43%) 33,670 (93.86%) 43,308 (93.95%) 37,305 (93.08%) 144,432 (93.80%) Screened in the previous year (all women) Yes NA 4,721 (13.16%) 6,104 (13.24%) 1,219 (3.04%) 12,044 (9.87%) No NA 31,152 (86.84%) 39,994 (86.76%) 38,861 (96.96%) 110,007 (90.13%) First-time screening within the past 3 years Yes NA NA 36,077 (78.26%) 35,364 (88.23%) 71,441 (82.89%) No NA NA 10,021 (21.74%) 4,716 (11.77%) 14,737 (17.10%) CI, confidence interval; cum, cumulative; HPV, human papillomavirus; NA, not applicable; VIA, visual inspection with acetic acid; VILI, visual inspection with Lugol’s iodine; y, year.
- Table 2
Impact of real-time OCR de-duplication on cervical cancer over-screening rates by age, screening strategy, and county cluster, 2021–2024
Item Pre-digital (%, n/N) Post-digital (%, n/N) Absolute delta (%, 95% CI)a P Adjusted Pb 2022 (1-y) 2023 (1-y) 2023 (2-y cum) 2024 (1-y-post) Age 35–44 15.57 (1,610/10,338) 14.97 (2,002/13,372) 23.18 (3,100/13,372) 0.18 (19/10,700) 15.06 (14.59, 15.52) < 0.001c < 0.001 45–54 12.03 (2,013/16,735) 13.34 (2,768/20,745) 22.15 (4,596/20,745) 0.20 (34/16,589) 12.55 (12.21, 12.90) < 0.001c < 0.001 55–64 7.57 (661/8,736) 8.80 (1,054/11,971) 16.67 (1,996/11,971) 0.13 (17/12,791) 8.15 (7.77, 8.53) < 0.001c < 0.001 Screening strategy HPV 6.29 (304/4,836) 3.85 (126/3,269) 12.97 (424/3,269) 0.00 (0/2,470) 5.31 (4.82, 5.79) < 0.001 < 0.001 HPV + cytology 1.62 (18/1,110) 1.33 (156/11,769) 3.62 (426/11,769) 0.33 (31/9,318) 1.02 (0.79, 1.25) < 0.001c < 0.001 Cytology 13.27 (3,962/29,863) 23.79 (2,399/10,083) 39.21 (3,954/10,083) NA NA NA NA HPV + VIA/VILI triage NA 8.05 (747/9,278) 16.74 (1,553/9,278) 0.19 (9/4,667) 7.86 (7.29, 8.43) < 0.001c < 0.001 HPV + cytology triage NA 20.50 (2,396/11,689) 28.53 (3,335/11,689) 0.14 (23/16,435) 20.36 (19.62, 21.09) < 0.001c < 0.001 HPV + cytology + VIA/VILI NA NA NA 0.10 (7/7,190) NA NA NA VIA + cytology triage NA NA NA NA NA NA NA Manual de-duplication cluster Zezhou County 7.60 (330/4,342) 1.56 (138/8,867) 4.79 (425/8,867) 0.06 (4/7,058) 3.49 (3.17, 3.81) < 0.001d < 0.001 Xiangyuan County 5.43 (325/5,990) 2.44 (128/5,246) 8.41 (441/5,246) 0.11 (6/5,646) 3.93 (3.55, 4.30) < 0.001c < 0.001 Xinping County 5.17 (213/4,117) 9.40 (747/7,943) 19.55 (1,553/7,943) 0.12 (10/8,506) 7.84 (7.35, 8.33) < 0.001c < 0.001 Yanting County 0.06 (4/6,920) 0.04 (3/7,003) 0.07 (5/7,003) 0.01 (1/7,621) 0.04 (−0.01, 0.08) 0.274d 1.000 Overall 4.08 (872/21,369) 3.50 (1,016/29,059) 8.34 (2,424/29,059) 0.07 (21/28,831) 3.67 (3.50, 3.84) < 0.001c < 0.001 No de-duplication cluster Shimian County 18.09 (724/4,002) 31.92 (1,722/5,395) 51.07 (2,755/5,395) 0.00 (0/3,180) 26.03 (25.14, 26.92) < 0.001d < 0.001 Mangshi 25.75 (2,688/10,438) 26.53 (3,086/11,634) 38.79 (4,513/11,634) 0.61 (49/8,069) 25.55 (24.95, 26.16) < 0.001c < 0.001 Overall 23.63 (3,412/14,440) 28.23 (4,808/17,029) 42.68 (7,268/17,029) 0.44 (49/11,249) 25.69 (25.18, 26.19) < 0.001c < 0.001 Overall 11.96 (4,284/35,809) 12.64 (5,824/46,088) 21.03 (9,692/46,088) 0.17 (70/40,080) 12.17 (11.94, 12.40) < 0.001c < 0.001 *Repeat screening within the ≤ 3-year guideline interval was defined as any woman who (i) had a documented negative cervical test in the relevant look-back window and (ii) underwent screening again in the index year. The look-back windows were 1 year for 2022 (records from 2021 only) and 2023 (records from 2022 only) and 2 years for 2023 (records from 2021–2022), and 2024 (records from 2023 after digital implementation). The over-screening rates were calculated as the number of such women (numerator) divided by the total number of negative (NEG) women screened in the same index year (denominator). aAbsolute delta was calculated as the overall over-screening rate in the pre-digital period minus that in the post-digital period. bAdjusted P-values were obtained using Bonferroni correction for 16 independent tests. cχ2 test comparing 2024 with pooled 2021–23 counts. dFisher exact test comparing 2024 with pooled 2021–23 counts due to expected cell counts < 5. CI, confidence interval; cum, cumulative; HPV, human papillomavirus; NA, not applicable; VIA, visual inspection with acetic acid; VILI, visual inspection with Lugol’s iodine; y, year.
CIN2+ Pb Adjusted Pc Cervical cancer P Adjusted Pc Pre-digital (%, n/N) Post-digital (%, n/N) Absolute deltaa (%, 95% CI) Pre-digital (%, n/N) Post-digital (%, n/N) Absolute delta (%, 95% CI) Overall 0.35 (275/78,958) 0.67 (505/75,020) 0.32 (0.25, 0.40) < 0.001 < 0.001 0.04 (34/78,958) 0.04 (31/75,020) −0.00 (−0.02, 0.02) 0.868b 1.000 Age group 35–44 0.35 (84/24,320) 0.63 (127/20,230) 0.28 (0.15, 0.41) < 0.001 < 0.001 0.05 (11/24,320) 0.02 (5/20,230) −0.02 (−0.05, 0.01) 0.255b 1.000 45–54 0.29 (109/36,985) 0.60 (194/32,314) 0.31 (0.20, 0.41) < 0.001 < 0.001 0.02 (8/36,985) 0.03 (10/32,314) 0.01 (−0.02, 0.03) 0.448b 1.000 55–64 0.46 (82/17,653) 0.82 (184/22,476) 0.35 (0.20, 0.51) < 0.001 < 0.001 0.08 (15/17,653) 0.07 (16/22,476) −0.01 (−0.07, 0.04) 0.622b 1.000 Site Zezhou County 0.30 (44/14,500) 0.53 (69/13,013) 0.23 (0.07, 0.38) 0.003 0.040 0.04 (6/14,500) 0.04 (5/13,013) −0.00 (−0.05, 0.04) 0.903b 1.000 Xiangyuan County 0.79 (88/11,171) 0.82 (89/10,893) 0.03 (−0.21, 0.26) 0.807 1.000 0.08 (9/11,171) 0.05 (5/10,893) −0.03 (−0.10, 0.03) 0.307b 1.000 Yanting County 0.48 (52/10,899) 1.02 (149/14,624) 0.54 (0.33, 0.75) < 0.001 < 0.001 0.06 (7/10,899) 0.02 (3/14,624) −0.04 (−0.10, 0.01) 0.110d 1.000 Shimian County 0.30 (32/10,782) 0.49 (26/5,351) 0.19 (−0.02, 0.40) 0.059 0.706 0.05 (5/10,782) 0.07 (4/5,351) 0.03 (−0.06, 0.11) 0.491d 1.000 Mangshi 0.15 (36/23,725) 0.27 (39/14,690) 0.11 (0.02, 0.21) 0.014 0.169 0.02 (4/23,725) 0.05 (7/14,690) 0.03 (−0.01, 0.07) 0.118d 1.000 Xinping County 0.29 (23/7,881) 0.81 (133/16,449) 0.52 (0.34, 0.70) < 0.001 < 0.001 0.04 (3/7,881) 0.04 (7/16,449) 0.00 (−0.05, 0.06) 1.000d 1.000 Strategy HPV 0.51 (48/9,358) 0.70 (40/5,739) 0.18 (−0.08, 0.44) 0.149 1.000 0.06 (6/9,358) 0.05 (3/5,739) −0.01 (−0.09, 0.07) 1.000d 1.000 HPV + VIA/VILI Triage NA (0/0) 0.83 (116/13,945) NA NA NA NA (0/0) 0.06 (9/13,945) NA (0/0) NA NA HPV + cytology 0.97 (22/2,276) 0.57 (114/20,019) −0.40 (−0.81, 0.02) 0.021 0.253 0.13 (3/2,276) 0.01 (3/20,019) −0.12 (−0.27, 0.03) 0.017d 0.201 HPV + cytology Triage NA (0/0) 0.62 (173/28,127) NA NA NA NA (0/0) 0.05 (14/28,127) NA (0/0) NA NA HPV + cytology + VIA/VILI NA (0/0) 0.86 (62/7,190) NA NA NA NA (0/0) 0.03 (2/7,190) NA (0/0) NA NA VIA + cytology Triage 0.18 (15/8,272) NA (0/0) NA NA NA 0.00 (0/8,272) NA (0/0) NA (0/0) NA NA Cytology 0.32 (190/59,052) NA (0/0) NA NA NA 0.04 (25/59,052) NA (0/0) NA (0/0) NA NA ↵aAbsolute delta was defined as the post-digital rate minus the pre-digital rate. Positive values indicate the improvement. bχ2 test comparing 2024 with pooled 2021–23 counts. cAdjusted p-values were obtained using Bonferroni correction for 12 independent tests. dFisher exact test comparing 2024 with pooled 2021–23 counts due to expected cell counts < 5. CI, confidence interval; VIA, visual inspection with acetic acid; VILI, visual inspection with Lugol’s iodine.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.